* Re: [PATCH v2] net: Require exact match for TCP socket lookups if dif is l3mdev
From: David Ahern @ 2016-10-14 12:21 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1476426789.5650.34.camel@edumazet-glaptop3.roam.corp.google.com>
On 10/14/16 12:33 AM, Eric Dumazet wrote:
> There is a catch here.
> TCP moves IP6CB() in a different location.
>
> Reference :
>
> 971f10eca186 ("tcp: better TCP_SKB_CB layout to reduce cache line misses")
thanks for the reference.
> Problem is that the lookup can happen from IP early demux, before TCP
> moved IP{6}CB around.
For TCP we only need the exact_dif match for listen sockets, so early demux does not apply.
>
> So you might need to let the caller pass IP6CB(skb)->flags (or
> TCP_SKB_CB(skb)->header.h6.flags ) instead of skb since
> inet6_exact_dif_match() does not know where to fetch the flags.
>
> Same issue for IPv4.
I'll update the match functions to pull from TCP_SKB_CB instead of IP6CB and make a note of the above.
Thanks for the review
^ permalink raw reply
* Re: [PATCH net-next 1/2] net: phy: Add Speed downshift set driver for Microsemi PHYs.
From: Andrew Lunn @ 2016-10-14 12:12 UTC (permalink / raw)
To: Raju Lakkaraju; +Cc: netdev, devicetree, f.fainelli, Allan.Nielsen
In-Reply-To: <1476445233-26524-2-git-send-email-Raju.Lakkaraju@microsemi.com>
On Fri, Oct 14, 2016 at 05:10:32PM +0530, Raju Lakkaraju wrote:
> From: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
>
> For operation in cabling environments that are incompatible with
> 1000BAST-T, VSC8531 device provides an automatic link speed
> downshift operation. When enabled, the device automatically changes
> its 1000BAST-T auto-negotiation to the next slower speed after
> a configured number of failed attempts at 1000BAST-T.
> This feature is useful in setting up in networks using older cable
> installations that include only pairs A and B, and not pairs C and D.
Any reason not to just turn this on by default when auto-neg is
enabled?
Andrew
^ permalink raw reply
* Re: Need help with mdiobus_register and phy
From: Andrew Lunn @ 2016-10-14 12:06 UTC (permalink / raw)
To: Timur Tabi; +Cc: netdev
In-Reply-To: <5800C3C7.60705@codeaurora.org>
On Fri, Oct 14, 2016 at 06:38:47AM -0500, Timur Tabi wrote:
> Andrew Lunn wrote:
> >Normally, a sleeping PHY does respond to MDIO. Otherwise, how do you
> >wake it?
> >
> >So i assume this phy has some other means to wake it. What is this
> >means?
>
> I'm guessing that someone has to call phy_resume() before/during the
> call to mdiobus_register, but I don't see how that's possible.
Please can you tell us what PHY which is, and how it is put to sleep
and woken up.
If the PHY cannot be woken up using MDIO, then maybe you need to look
at the mdio bus reset call?
Andrew
^ permalink raw reply
* Re: [PATCH net-next 2/2] net: phy: Add Fast Link Failure - 2 set driver for Microsemi PHYs.
From: Andrew Lunn @ 2016-10-14 12:02 UTC (permalink / raw)
To: Raju Lakkaraju; +Cc: netdev, devicetree, f.fainelli, Allan.Nielsen
In-Reply-To: <1476445233-26524-3-git-send-email-Raju.Lakkaraju@microsemi.com>
> On Fri, Oct 14, 2016 at 05:10:33PM +0530, Raju Lakkaraju wrote:
> From: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
>
> VSC8531 Fast Link Failure 2 feature enables the PHY to indicate the
> onset of a potential link failure in < 100 usec for 100BASE-TX
> operation. FLF2 is supported through the MDINT (active low) pin.
Is the MDINT pin specific to this feature, or a general interrupt pin?
Device tree is used to describe the hardware. It should not really
describe software or configuration. But the borders are a bit
fluffly. Signal edge rates is near to hardware. This is a lot more
towards configuration. So i'm not sure a device tree property is the
correct way to describe this.
This is also a feature i know other PHYs support. The Marvell PHY has
a "Metro Ethernet" extension which allows it to report link failures
for 1000BASE-T in 10, 20 or 40ms, instead of the usual 750ms. So we
need a generic solution other PHYs can implement.
As with cable testing, i think it should be an ethtool option.
Andrew
^ permalink raw reply
* [PATCH net-next 0/2] net: phy: Add Downshift, FLF2 drivers for Microsemi
From: Raju Lakkaraju @ 2016-10-14 11:40 UTC (permalink / raw)
To: netdev, devicetree; +Cc: f.fainelli, Allan.Nielsen, andrew, Raju Lakkaraju
From: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
This series adds support to the Speed downshift, Fast Link Failure 2,
set drivers for Microsemi PHYs.
Patch 1/4: Link Speed downshift:
For operation in cabling environments that are incompatible with
1000BAST-T, VSC8531 device provides an automatic link speed
downshift operation. When enabled, the device automatically changes
its 1000BAST-T auto-negotiation to the next slower speed after
a set number of failed attempts at 1000BAST-T.
This feature is useful in setting up in networks using older cable
installations that include only pairs A and B, and not pairs C and D.
Patch 2/4: Fast Link Failure 2:
VSC8531 Fast Link Failure 2 feature enables the PHY to indicate the
onset of a potential link failure in < 100 usec for 100BASE-TX
operation. FLF2 is supported through the MDINT (active low) pin.
All these features tested on Beaglebone Black with VSC 8531 PHY.
Raju Lakkaraju (2):
net: phy: Add Speed downshift set driver for Microsemi PHYs.
net: phy: Add Fast Link Failure - 2 set driver for Microsemi PHYs.
.../devicetree/bindings/net/mscc-phy-vsc8531.txt | 12 +++
drivers/net/phy/mscc.c | 120 ++++++++++++++++++++-
2 files changed, 131 insertions(+), 1 deletion(-)
--
2.7.4
^ permalink raw reply
* [PATCH net-next 2/2] net: phy: Add Fast Link Failure - 2 set driver for Microsemi PHYs.
From: Raju Lakkaraju @ 2016-10-14 11:40 UTC (permalink / raw)
To: netdev, devicetree
Cc: f.fainelli, Allan.Nielsen, andrew, Raju Lakkaraju,
Allan W . Nielsen
In-Reply-To: <1476445233-26524-1-git-send-email-Raju.Lakkaraju@microsemi.com>
From: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
VSC8531 Fast Link Failure 2 feature enables the PHY to indicate the
onset of a potential link failure in < 100 usec for 100BASE-TX
operation. FLF2 is supported through the MDINT (active low) pin.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
Signed-off-by: Allan W. Nielsen <allan.nielsen@microsemi.com>
---
.../devicetree/bindings/net/mscc-phy-vsc8531.txt | 6 +++
drivers/net/phy/mscc.c | 45 ++++++++++++++++++++++
2 files changed, 51 insertions(+)
diff --git a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt
index 062d115..472fc68 100644
--- a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt
+++ b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt
@@ -32,6 +32,11 @@ Optional properties:
after a 'downshift-cnt' of failed attempts at
1000BAST-T. Allowed values: 0, 2, 3, 4, 5.
0 is default and will disable downshifting.
+- flf2 : Fast Link Failure 2 (FLF2) feature enables the PHY
+ to indicate the onset of a potential link failure in
+ < 100 usec for 100BASE-TX operation. FLF2 is
+ supported through the MDINT (active low) pin.
+ Default will be disable flf2.
Table: 1 - Edge rate change
----------------------------------------------------------------|
@@ -66,4 +71,5 @@ Example:
vsc8531,vddmac = <3300>;
vsc8531,edge-slowdown = <7>;
vsc8531,downshift-cnt = <3>;
+ vsc8531,flf2;
};
diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c
index e87d9f0..57bd628 100644
--- a/drivers/net/phy/mscc.c
+++ b/drivers/net/phy/mscc.c
@@ -57,6 +57,7 @@ enum rgmii_rx_clock_delay {
/* Extended Page 2 Registers */
#define MSCC_PHY_RGMII_CNTL 20
+#define FLF2_ENABLE 0x8000
#define RGMII_RX_CLK_DELAY_MASK 0x0070
#define RGMII_RX_CLK_DELAY_POS 4
@@ -83,6 +84,7 @@ enum rgmii_rx_clock_delay {
struct vsc8531_private {
int rate_magic;
u8 downshift_magic;
+ bool flf2; /* Fast Link Failure-2 Enable/Disable */
};
#ifdef CONFIG_OF_MDIO
@@ -107,6 +109,33 @@ static int vsc85xx_phy_page_set(struct phy_device *phydev, u8 page)
return rc;
}
+static int vsc85xx_flf2_set(struct phy_device *phydev, bool op)
+{
+ int rc;
+ u16 reg_val;
+
+ mutex_lock(&phydev->lock);
+ rc = vsc85xx_phy_page_set(phydev, MSCC_PHY_PAGE_EXTENDED_2);
+ if (rc != 0)
+ goto out_unlock;
+
+ reg_val = phy_read(phydev, MSCC_PHY_RGMII_CNTL);
+ if (op)
+ reg_val |= FLF2_ENABLE;
+ else
+ reg_val &= ~FLF2_ENABLE;
+ rc = phy_write(phydev, MSCC_PHY_RGMII_CNTL, reg_val);
+ if (rc != 0)
+ goto out_unlock;
+
+ rc = vsc85xx_phy_page_set(phydev, MSCC_PHY_PAGE_STANDARD);
+
+out_unlock:
+ mutex_unlock(&phydev->lock);
+
+ return rc;
+}
+
static int vsc85xx_downshift_set(struct phy_device *phydev, u8 magic)
{
int rc;
@@ -412,6 +441,10 @@ static int vsc85xx_config_init(struct phy_device *phydev)
if (rc)
return rc;
+ rc = vsc85xx_flf2_set(phydev, vsc8531->flf2);
+ if (rc)
+ return rc;
+
rc = genphy_config_init(phydev);
return rc;
@@ -449,6 +482,11 @@ static int vsc85xx_probe(struct phy_device *phydev)
int rate_magic;
int downshift_magic;
struct vsc8531_private *vsc8531;
+ struct device *dev = &phydev->mdio.dev;
+ struct device_node *of_node = dev->of_node;
+
+ if (!of_node)
+ return -ENODEV;
rate_magic = vsc85xx_edge_rate_magic_get(phydev);
if (rate_magic < 0)
@@ -466,6 +504,13 @@ static int vsc85xx_probe(struct phy_device *phydev)
vsc8531->rate_magic = rate_magic;
vsc8531->downshift_magic = downshift_magic;
+#ifdef CONFIG_OF_MDIO
+ /* Fast Link Failure 2 */
+ vsc8531->flf2 = of_property_read_bool(of_node, "vsc8531,flf2");
+#else
+ vsc8531->flf2 = 0;
+#endif
+
return 0;
}
--
2.7.4
^ permalink raw reply related
* [PATCH 4.4 01/21] time: Add cycles to nanoseconds translation
From: Greg Kroah-Hartman @ 2016-10-14 11:40 UTC (permalink / raw)
To: linux-kernel
Cc: Greg Kroah-Hartman, stable, Prarit Bhargava, Richard Cochran,
Thomas Gleixner, Ingo Molnar, Andy Lutomirski, kevin.b.stanton,
kevin.j.clarke, hpa, jeffrey.t.kirsher, netdev,
Christopher S. Hall, John Stultz
In-Reply-To: <20161014114024.667943585@linuxfoundation.org>
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christopher S. Hall <christopher.s.hall@intel.com>
commit 6bd58f09e1d8cc6c50a824c00bf0d617919986a1 upstream.
The timekeeping code does not currently provide a way to translate
externally provided clocksource cycles to system time. The cycle count
is always provided by the result clocksource read() method internal to
the timekeeping code. The added function timekeeping_cycles_to_ns()
calculated a nanosecond value from a cycle count that can be added to
tk_read_base.base value yielding the current system time. This allows
clocksource cycle values external to the timekeeping code to provide a
cycle count that can be transformed to system time.
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: kevin.b.stanton@intel.com
Cc: kevin.j.clarke@intel.com
Cc: hpa@zytor.com
Cc: jeffrey.t.kirsher@intel.com
Cc: netdev@vger.kernel.org
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Christopher S. Hall <christopher.s.hall@intel.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
kernel/time/timekeeping.c | 25 +++++++++++++++++++++----
1 file changed, 21 insertions(+), 4 deletions(-)
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -298,17 +298,34 @@ u32 (*arch_gettimeoffset)(void) = defaul
static inline u32 arch_gettimeoffset(void) { return 0; }
#endif
+static inline s64 timekeeping_delta_to_ns(struct tk_read_base *tkr,
+ cycle_t delta)
+{
+ s64 nsec;
+
+ nsec = delta * tkr->mult + tkr->xtime_nsec;
+ nsec >>= tkr->shift;
+
+ /* If arch requires, add in get_arch_timeoffset() */
+ return nsec + arch_gettimeoffset();
+}
+
static inline s64 timekeeping_get_ns(struct tk_read_base *tkr)
{
cycle_t delta;
- s64 nsec;
delta = timekeeping_get_delta(tkr);
+ return timekeeping_delta_to_ns(tkr, delta);
+}
- nsec = (delta * tkr->mult + tkr->xtime_nsec) >> tkr->shift;
+static inline s64 timekeeping_cycles_to_ns(struct tk_read_base *tkr,
+ cycle_t cycles)
+{
+ cycle_t delta;
- /* If arch requires, add in get_arch_timeoffset() */
- return nsec + arch_gettimeoffset();
+ /* calculate the delta since the last update_wall_time */
+ delta = clocksource_delta(cycles, tkr->cycle_last, tkr->mask);
+ return timekeeping_delta_to_ns(tkr, delta);
}
/**
^ permalink raw reply
* [PATCH net-next 1/2] net: phy: Add Speed downshift set driver for Microsemi PHYs.
From: Raju Lakkaraju @ 2016-10-14 11:40 UTC (permalink / raw)
To: netdev-u79uwXL29TY76Z2rM5mHXA, devicetree-u79uwXL29TY76Z2rM5mHXA
Cc: f.fainelli-Re5JQEeQqe8AvxtiuMwx3w,
Allan.Nielsen-dzo6w/eZyo2tG0bUXCXiUA, andrew-g2DYL2Zd6BY,
Raju Lakkaraju, Allan W . Nielsen
In-Reply-To: <1476445233-26524-1-git-send-email-Raju.Lakkaraju-dzo6w/eZyo2tG0bUXCXiUA@public.gmane.org>
From: Raju Lakkaraju <Raju.Lakkaraju-dzo6w/eZyo2tG0bUXCXiUA@public.gmane.org>
For operation in cabling environments that are incompatible with
1000BAST-T, VSC8531 device provides an automatic link speed
downshift operation. When enabled, the device automatically changes
its 1000BAST-T auto-negotiation to the next slower speed after
a configured number of failed attempts at 1000BAST-T.
This feature is useful in setting up in networks using older cable
installations that include only pairs A and B, and not pairs C and D.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju-dzo6w/eZyo2tG0bUXCXiUA@public.gmane.org>
Signed-off-by: Allan W. Nielsen <allan.nielsen-dzo6w/eZyo2tG0bUXCXiUA@public.gmane.org>
---
.../devicetree/bindings/net/mscc-phy-vsc8531.txt | 6 ++
drivers/net/phy/mscc.c | 75 +++++++++++++++++++++-
2 files changed, 80 insertions(+), 1 deletion(-)
diff --git a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt
index bdefefc6..062d115 100644
--- a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt
+++ b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt
@@ -27,6 +27,11 @@ Optional properties:
'vddmac'.
Default value is 0%.
Ref: Table:1 - Edge rate change (below).
+- downshift-cnt : When enabled, the device automatically changes its
+ 1000BAST-T auto-negotiation to the next slower speed
+ after a 'downshift-cnt' of failed attempts at
+ 1000BAST-T. Allowed values: 0, 2, 3, 4, 5.
+ 0 is default and will disable downshifting.
Table: 1 - Edge rate change
----------------------------------------------------------------|
@@ -60,4 +65,5 @@ Example:
compatible = "ethernet-phy-id0007.0570";
vsc8531,vddmac = <3300>;
vsc8531,edge-slowdown = <7>;
+ vsc8531,downshift-cnt = <3>;
};
diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c
index 43a7545..e87d9f0 100644
--- a/drivers/net/phy/mscc.c
+++ b/drivers/net/phy/mscc.c
@@ -46,8 +46,15 @@ enum rgmii_rx_clock_delay {
#define MSCC_EXT_PAGE_ACCESS 31
#define MSCC_PHY_PAGE_STANDARD 0x0000 /* Standard registers */
+#define MSCC_PHY_PAGE_EXTENDED 0x0001 /* Extended registers */
#define MSCC_PHY_PAGE_EXTENDED_2 0x0002 /* Extended reg - page 2 */
+/* Extended Page 1 Registers */
+#define MSCC_PHY_ACTIPHY_CNTL 20
+#define DOWNSHIFT_CNTL_MASK 0x000C
+#define DOWNSHIFT_EN 0x0010
+#define DOWNSHIFT_CNTL_POS 2
+
/* Extended Page 2 Registers */
#define MSCC_PHY_RGMII_CNTL 20
#define RGMII_RX_CLK_DELAY_MASK 0x0070
@@ -75,6 +82,7 @@ enum rgmii_rx_clock_delay {
struct vsc8531_private {
int rate_magic;
+ u8 downshift_magic;
};
#ifdef CONFIG_OF_MDIO
@@ -99,6 +107,31 @@ static int vsc85xx_phy_page_set(struct phy_device *phydev, u8 page)
return rc;
}
+static int vsc85xx_downshift_set(struct phy_device *phydev, u8 magic)
+{
+ int rc;
+ u16 reg_val;
+
+ mutex_lock(&phydev->lock);
+ rc = vsc85xx_phy_page_set(phydev, MSCC_PHY_PAGE_EXTENDED);
+ if (rc != 0)
+ goto out_unlock;
+
+ reg_val = phy_read(phydev, MSCC_PHY_ACTIPHY_CNTL);
+ reg_val &= ~(DOWNSHIFT_CNTL_MASK);
+ reg_val |= magic;
+ rc = phy_write(phydev, MSCC_PHY_ACTIPHY_CNTL, reg_val);
+ if (rc != 0)
+ goto out_unlock;
+
+ rc = vsc85xx_phy_page_set(phydev, MSCC_PHY_PAGE_STANDARD);
+
+out_unlock:
+ mutex_unlock(&phydev->lock);
+
+ return rc;
+}
+
static int vsc85xx_wol_set(struct phy_device *phydev,
struct ethtool_wolinfo *wol)
{
@@ -239,11 +272,42 @@ static int vsc85xx_edge_rate_magic_get(struct phy_device *phydev)
return -EINVAL;
}
+
+static int vsc85xx_downshift_magic_get(struct phy_device *phydev)
+{
+ int rc;
+ u8 ds;
+ struct device *dev = &phydev->mdio.dev;
+ struct device_node *of_node = dev->of_node;
+
+ if (!of_node)
+ return -ENODEV;
+
+ rc = of_property_read_u8(of_node, "vsc8531,downshift-cnt", &ds);
+ if ((rc == -EINVAL) || (ds == 0))
+ return 0;
+ if (ds == 1 || ds > 5) {
+ phydev_err(phydev, "Invalid downshift count\n");
+ return -EINVAL;
+ }
+
+ /* ds is either 2,3,4 or 5 */
+ ds -= 2;
+ ds <<= DOWNSHIFT_CNTL_POS;
+ ds |= DOWNSHIFT_EN;
+
+ return ds;
+}
#else
static int vsc85xx_edge_rate_magic_get(struct phy_device *phydev)
{
return 0;
}
+
+static int vsc85xx_downshift_magic_get(struct phy_device *phydev)
+{
+ return 0;
+}
#endif /* CONFIG_OF_MDIO */
static int vsc85xx_edge_rate_cntl_set(struct phy_device *phydev, u8 edge_rate)
@@ -344,6 +408,10 @@ static int vsc85xx_config_init(struct phy_device *phydev)
if (rc)
return rc;
+ rc = vsc85xx_downshift_set(phydev, vsc8531->downshift_magic);
+ if (rc)
+ return rc;
+
rc = genphy_config_init(phydev);
return rc;
@@ -379,19 +447,24 @@ static int vsc85xx_config_intr(struct phy_device *phydev)
static int vsc85xx_probe(struct phy_device *phydev)
{
int rate_magic;
+ int downshift_magic;
struct vsc8531_private *vsc8531;
rate_magic = vsc85xx_edge_rate_magic_get(phydev);
if (rate_magic < 0)
return rate_magic;
+ downshift_magic = vsc85xx_downshift_magic_get(phydev);
+ if (downshift_magic < 0)
+ return downshift_magic;
+
vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL);
if (!vsc8531)
return -ENOMEM;
phydev->priv = vsc8531;
-
vsc8531->rate_magic = rate_magic;
+ vsc8531->downshift_magic = downshift_magic;
return 0;
}
--
2.7.4
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH] r8169: set coherent DMA mask as well as streaming DMA mask
From: Ard Biesheuvel @ 2016-10-14 11:39 UTC (permalink / raw)
To: romieu, nic_swsd, netdev, linux-kernel, davem; +Cc: Ard Biesheuvel
PCI devices that are 64-bit DMA capable should set the coherent
DMA mask as well as the streaming DMA mask. On some architectures,
these are managed separately, and so the coherent DMA mask will be
left at its default value of 32 if it is not set explicitly. This
results in errors such as
r8169 Gigabit Ethernet driver 2.3LK-NAPI loaded
hwdev DMA mask = 0x00000000ffffffff, dev_addr = 0x00000080fbfff000
swiotlb: coherent allocation failed for device 0000:02:00.0 size=4096
CPU: 0 PID: 1062 Comm: systemd-udevd Not tainted 4.8.0+ #35
Hardware name: AMD Seattle/Seattle, BIOS 10:53:24 Oct 13 2016
on systems without memory that is 32-bit addressable by PCI devices.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
---
drivers/net/ethernet/realtek/r8169.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index e55638c7505a..04957a36b11f 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -8273,7 +8273,8 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
if ((sizeof(dma_addr_t) > 4) &&
(use_dac == 1 || (use_dac == -1 && pci_is_pcie(pdev) &&
tp->mac_version >= RTL_GIGA_MAC_VER_18)) &&
- !pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
+ !pci_set_dma_mask(pdev, DMA_BIT_MASK(64)) &&
+ !pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64))) {
/* CPlusCmd Dual Access Cycle is only needed for non-PCIe */
if (!pci_is_pcie(pdev))
@@ -8281,6 +8282,8 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
dev->features |= NETIF_F_HIGHDMA;
} else {
rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+ if (!rc)
+ rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
if (rc < 0) {
netif_err(tp, probe, dev, "DMA configuration failed\n");
goto err_out_unmap_4;
--
2.7.4
^ permalink raw reply related
* [PATCH net-next v2 6/6] fjes: Update fjes driver version : 1.2
From: Taku Izumi @ 2016-10-14 11:28 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
In-Reply-To: <1476444345-28673-1-git-send-email-izumi.taku@jp.fujitsu.com>
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
---
drivers/net/fjes/fjes_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c
index 359e7a5..f36eb4a 100644
--- a/drivers/net/fjes/fjes_main.c
+++ b/drivers/net/fjes/fjes_main.c
@@ -30,7 +30,7 @@
#include "fjes_trace.h"
#define MAJ 1
-#define MIN 1
+#define MIN 2
#define DRV_VERSION __stringify(MAJ) "." __stringify(MIN)
#define DRV_NAME "fjes"
char fjes_driver_name[] = DRV_NAME;
--
2.6.6
^ permalink raw reply related
* [PATCH net-next v2 5/6] fjes: Add debugfs entry for EP status information in fjes driver
From: Taku Izumi @ 2016-10-14 11:28 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
In-Reply-To: <1476444345-28673-1-git-send-email-izumi.taku@jp.fujitsu.com>
This patch adds debugfs entry to show EP status information.
You can get each EP's status information like the following:
# cat /sys/kernel/debug/fjes/fjes.0/status
EPID STATUS SAME_ZONE CONNECTED
ep0 shared Y Y
ep1 - - -
ep2 unshared N N
ep3 unshared N N
ep4 unshared N N
ep5 unshared N N
ep6 unshared N N
ep7 unshared N N
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
---
drivers/net/fjes/Makefile | 2 +-
drivers/net/fjes/fjes.h | 16 ++++++
drivers/net/fjes/fjes_debugfs.c | 117 ++++++++++++++++++++++++++++++++++++++++
drivers/net/fjes/fjes_main.c | 12 ++++-
4 files changed, 145 insertions(+), 2 deletions(-)
create mode 100644 drivers/net/fjes/fjes_debugfs.c
diff --git a/drivers/net/fjes/Makefile b/drivers/net/fjes/Makefile
index 6705d1b..bc47b35 100644
--- a/drivers/net/fjes/Makefile
+++ b/drivers/net/fjes/Makefile
@@ -27,4 +27,4 @@
obj-$(CONFIG_FUJITSU_ES) += fjes.o
-fjes-objs := fjes_main.o fjes_hw.o fjes_ethtool.o fjes_trace.o
+fjes-objs := fjes_main.o fjes_hw.o fjes_ethtool.o fjes_trace.o fjes_debugfs.o
diff --git a/drivers/net/fjes/fjes.h b/drivers/net/fjes/fjes.h
index a592fe2..0372be3 100644
--- a/drivers/net/fjes/fjes.h
+++ b/drivers/net/fjes/fjes.h
@@ -66,6 +66,10 @@ struct fjes_adapter {
bool interrupt_watch_enable;
struct fjes_hw hw;
+
+#ifdef CONFIG_DEBUG_FS
+ struct dentry *dbg_adapter;
+#endif
};
extern char fjes_driver_name[];
@@ -74,4 +78,16 @@ extern const u32 fjes_support_mtu[];
void fjes_set_ethtool_ops(struct net_device *);
+#ifdef CONFIG_DEBUG_FS
+void fjes_dbg_adapter_init(struct fjes_adapter *adapter);
+void fjes_dbg_adapter_exit(struct fjes_adapter *adapter);
+void fjes_dbg_init(void);
+void fjes_dbg_exit(void);
+#else
+static inline void fjes_dbg_adapter_init(struct fjes_adapter *adapter) {}
+static inline void fjes_dbg_adapter_exit(struct fjes_adapter *adapter) {}
+static inline void fjes_dbg_init(void) {}
+static inline void fjes_dbg_exit(void) {}
+#endif /* CONFIG_DEBUG_FS */
+
#endif /* FJES_H_ */
diff --git a/drivers/net/fjes/fjes_debugfs.c b/drivers/net/fjes/fjes_debugfs.c
new file mode 100644
index 0000000..30052eb
--- /dev/null
+++ b/drivers/net/fjes/fjes_debugfs.c
@@ -0,0 +1,117 @@
+/*
+ * FUJITSU Extended Socket Network Device driver
+ * Copyright (c) 2015-2016 FUJITSU LIMITED
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ */
+
+/* debugfs support for fjes driver */
+
+#ifdef CONFIG_DEBUG_FS
+
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/platform_device.h>
+
+#include "fjes.h"
+
+static struct dentry *fjes_debug_root;
+
+static const char * const ep_status_string[] = {
+ "unshared",
+ "shared",
+ "waiting",
+ "complete",
+};
+
+static int fjes_dbg_status_show(struct seq_file *m, void *v)
+{
+ struct fjes_adapter *adapter = m->private;
+ struct fjes_hw *hw = &adapter->hw;
+ int max_epid = hw->max_epid;
+ int my_epid = hw->my_epid;
+ int epidx;
+
+ seq_puts(m, "EPID\tSTATUS SAME_ZONE CONNECTED\n");
+ for (epidx = 0; epidx < max_epid; epidx++) {
+ if (epidx == my_epid) {
+ seq_printf(m, "ep%d\t%-16c %-16c %-16c\n",
+ epidx, '-', '-', '-');
+ } else {
+ seq_printf(m, "ep%d\t%-16s %-16c %-16c\n",
+ epidx,
+ ep_status_string[fjes_hw_get_partner_ep_status(hw, epidx)],
+ fjes_hw_epid_is_same_zone(hw, epidx) ? 'Y' : 'N',
+ fjes_hw_epid_is_shared(hw->hw_info.share, epidx) ? 'Y' : 'N');
+ }
+ }
+
+ return 0;
+}
+
+static int fjes_dbg_status_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, fjes_dbg_status_show, inode->i_private);
+}
+
+static const struct file_operations fjes_dbg_status_fops = {
+ .owner = THIS_MODULE,
+ .open = fjes_dbg_status_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+void fjes_dbg_adapter_init(struct fjes_adapter *adapter)
+{
+ const char *name = dev_name(&adapter->plat_dev->dev);
+ struct dentry *pfile;
+
+ adapter->dbg_adapter = debugfs_create_dir(name, fjes_debug_root);
+ if (!adapter->dbg_adapter) {
+ dev_err(&adapter->plat_dev->dev,
+ "debugfs entry for %s failed\n", name);
+ return;
+ }
+
+ pfile = debugfs_create_file("status", 0444, adapter->dbg_adapter,
+ adapter, &fjes_dbg_status_fops);
+ if (!pfile)
+ dev_err(&adapter->plat_dev->dev,
+ "debugfs status for %s failed\n", name);
+}
+
+void fjes_dbg_adapter_exit(struct fjes_adapter *adapter)
+{
+ debugfs_remove_recursive(adapter->dbg_adapter);
+ adapter->dbg_adapter = NULL;
+}
+
+void fjes_dbg_init(void)
+{
+ fjes_debug_root = debugfs_create_dir(fjes_driver_name, NULL);
+ if (!fjes_debug_root)
+ pr_info("init of debugfs failed\n");
+}
+
+void fjes_dbg_exit(void)
+{
+ debugfs_remove_recursive(fjes_debug_root);
+ fjes_debug_root = NULL;
+}
+
+#endif /* CONFIG_DEBUG_FS */
diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c
index 4c135ae..359e7a5 100644
--- a/drivers/net/fjes/fjes_main.c
+++ b/drivers/net/fjes/fjes_main.c
@@ -1257,6 +1257,8 @@ static int fjes_probe(struct platform_device *plat_dev)
netif_carrier_off(netdev);
+ fjes_dbg_adapter_init(adapter);
+
return 0;
err_hw_exit:
@@ -1274,6 +1276,8 @@ static int fjes_remove(struct platform_device *plat_dev)
struct fjes_adapter *adapter = netdev_priv(netdev);
struct fjes_hw *hw = &adapter->hw;
+ fjes_dbg_adapter_exit(adapter);
+
cancel_delayed_work_sync(&adapter->interrupt_watch_task);
cancel_work_sync(&adapter->unshare_watch_task);
cancel_work_sync(&adapter->raise_intr_rxdata_task);
@@ -1478,9 +1482,13 @@ static int __init fjes_init_module(void)
pr_info("%s - version %s - %s\n",
fjes_driver_string, fjes_driver_version, fjes_copyright);
+ fjes_dbg_init();
+
result = platform_driver_register(&fjes_driver);
- if (result < 0)
+ if (result < 0) {
+ fjes_dbg_exit();
return result;
+ }
result = acpi_bus_register_driver(&fjes_acpi_driver);
if (result < 0)
@@ -1490,6 +1498,7 @@ static int __init fjes_init_module(void)
fail_acpi_driver:
platform_driver_unregister(&fjes_driver);
+ fjes_dbg_exit();
return result;
}
@@ -1500,6 +1509,7 @@ static void __exit fjes_exit_module(void)
{
acpi_bus_unregister_driver(&fjes_acpi_driver);
platform_driver_unregister(&fjes_driver);
+ fjes_dbg_exit();
}
module_exit(fjes_exit_module);
--
2.6.6
^ permalink raw reply related
* Re: Need help with mdiobus_register and phy
From: Timur Tabi @ 2016-10-14 11:38 UTC (permalink / raw)
To: Andrew Lunn; +Cc: netdev
In-Reply-To: <20161014040641.GE5822@lunn.ch>
Andrew Lunn wrote:
> Normally, a sleeping PHY does respond to MDIO. Otherwise, how do you
> wake it?
>
> So i assume this phy has some other means to wake it. What is this
> means?
I'm guessing that someone has to call phy_resume() before/during the
call to mdiobus_register, but I don't see how that's possible.
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the
Code Aurora Forum, hosted by The Linux Foundation.
^ permalink raw reply
* [PATCH net-next v2 4/6] fjes: ethtool -w and -W support for fjes driver
From: Taku Izumi @ 2016-10-14 11:27 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
In-Reply-To: <1476444345-28673-1-git-send-email-izumi.taku@jp.fujitsu.com>
This patch adds implementation of supporting
ethtool -w and -W for fjes driver.
You can enable and disable firmware debug mode by
using ethtool -W, and also retrieve firmware
activity information by using ethtool -w.
This is useful for debugging.
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
---
drivers/net/fjes/fjes_ethtool.c | 63 ++++++++++++++++++
drivers/net/fjes/fjes_hw.c | 137 ++++++++++++++++++++++++++++++++++++++++
drivers/net/fjes/fjes_hw.h | 15 +++++
drivers/net/fjes/fjes_trace.h | 69 ++++++++++++++++++++
4 files changed, 284 insertions(+)
diff --git a/drivers/net/fjes/fjes_ethtool.c b/drivers/net/fjes/fjes_ethtool.c
index 68ef287..6575f88 100644
--- a/drivers/net/fjes/fjes_ethtool.c
+++ b/drivers/net/fjes/fjes_ethtool.c
@@ -235,6 +235,66 @@ static void fjes_get_regs(struct net_device *netdev,
regs_buff[36] = rd32(XSCT_ICTL);
}
+static int fjes_set_dump(struct net_device *netdev, struct ethtool_dump *dump)
+{
+ struct fjes_adapter *adapter = netdev_priv(netdev);
+ struct fjes_hw *hw = &adapter->hw;
+ int ret = 0;
+
+ if (dump->flag) {
+ if (hw->debug_mode)
+ return -EPERM;
+
+ hw->debug_mode = dump->flag;
+
+ /* enable debug mode */
+ mutex_lock(&hw->hw_info.lock);
+ ret = fjes_hw_start_debug(hw);
+ mutex_unlock(&hw->hw_info.lock);
+
+ if (ret)
+ hw->debug_mode = 0;
+ } else {
+ if (!hw->debug_mode)
+ return -EPERM;
+
+ /* disable debug mode */
+ mutex_lock(&hw->hw_info.lock);
+ ret = fjes_hw_stop_debug(hw);
+ mutex_unlock(&hw->hw_info.lock);
+ }
+
+ return ret;
+}
+
+static int fjes_get_dump_flag(struct net_device *netdev,
+ struct ethtool_dump *dump)
+{
+ struct fjes_adapter *adapter = netdev_priv(netdev);
+ struct fjes_hw *hw = &adapter->hw;
+
+ dump->len = hw->hw_info.trace_size;
+ dump->version = 1;
+ dump->flag = hw->debug_mode;
+
+ return 0;
+}
+
+static int fjes_get_dump_data(struct net_device *netdev,
+ struct ethtool_dump *dump, void *buf)
+{
+ struct fjes_adapter *adapter = netdev_priv(netdev);
+ struct fjes_hw *hw = &adapter->hw;
+ int ret = 0;
+
+ if (hw->hw_info.trace)
+ memcpy(buf, hw->hw_info.trace, hw->hw_info.trace_size);
+ else
+ ret = -EPERM;
+
+ return ret;
+}
+
static const struct ethtool_ops fjes_ethtool_ops = {
.get_settings = fjes_get_settings,
.get_drvinfo = fjes_get_drvinfo,
@@ -243,6 +303,9 @@ static const struct ethtool_ops fjes_ethtool_ops = {
.get_sset_count = fjes_get_sset_count,
.get_regs = fjes_get_regs,
.get_regs_len = fjes_get_regs_len,
+ .set_dump = fjes_set_dump,
+ .get_dump_flag = fjes_get_dump_flag,
+ .get_dump_data = fjes_get_dump_data,
};
void fjes_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/fjes/fjes_hw.c b/drivers/net/fjes/fjes_hw.c
index dba59dc..9c652c0 100644
--- a/drivers/net/fjes/fjes_hw.c
+++ b/drivers/net/fjes/fjes_hw.c
@@ -343,6 +343,9 @@ int fjes_hw_init(struct fjes_hw *hw)
ret = fjes_hw_setup(hw);
+ hw->hw_info.trace = vzalloc(FJES_DEBUG_BUFFER_SIZE);
+ hw->hw_info.trace_size = FJES_DEBUG_BUFFER_SIZE;
+
return ret;
}
@@ -351,6 +354,18 @@ void fjes_hw_exit(struct fjes_hw *hw)
int ret;
if (hw->base) {
+
+ if (hw->debug_mode) {
+ /* disable debug mode */
+ mutex_lock(&hw->hw_info.lock);
+ fjes_hw_stop_debug(hw);
+ mutex_unlock(&hw->hw_info.lock);
+ }
+ vfree(hw->hw_info.trace);
+ hw->hw_info.trace = NULL;
+ hw->hw_info.trace_size = 0;
+ hw->debug_mode = 0;
+
ret = fjes_hw_reset(hw);
if (ret)
pr_err("%s: reset error", __func__);
@@ -1175,3 +1190,125 @@ static void fjes_hw_epstop_task(struct work_struct *work)
}
}
}
+
+int fjes_hw_start_debug(struct fjes_hw *hw)
+{
+ union fjes_device_command_req *req_buf = hw->hw_info.req_buf;
+ union fjes_device_command_res *res_buf = hw->hw_info.res_buf;
+ enum fjes_dev_command_response_e ret;
+ int page_count;
+ int result = 0;
+ void *addr;
+ int i;
+
+ if (!hw->hw_info.trace)
+ return -EPERM;
+ memset(hw->hw_info.trace, 0, FJES_DEBUG_BUFFER_SIZE);
+
+ memset(req_buf, 0, hw->hw_info.req_buf_size);
+ memset(res_buf, 0, hw->hw_info.res_buf_size);
+
+ req_buf->start_trace.length =
+ FJES_DEV_COMMAND_START_DBG_REQ_LEN(hw->hw_info.trace_size);
+ req_buf->start_trace.mode = hw->debug_mode;
+ req_buf->start_trace.buffer_len = hw->hw_info.trace_size;
+ page_count = hw->hw_info.trace_size / FJES_DEBUG_PAGE_SIZE;
+ for (i = 0; i < page_count; i++) {
+ addr = ((u8 *)hw->hw_info.trace) + i * FJES_DEBUG_PAGE_SIZE;
+ req_buf->start_trace.buffer[i] =
+ (__le64)(page_to_phys(vmalloc_to_page(addr)) +
+ offset_in_page(addr));
+ }
+
+ res_buf->start_trace.length = 0;
+ res_buf->start_trace.code = 0;
+
+ trace_fjes_hw_start_debug_req(req_buf);
+ ret = fjes_hw_issue_request_command(hw, FJES_CMD_REQ_START_DEBUG);
+ trace_fjes_hw_start_debug(res_buf);
+
+ if (res_buf->start_trace.length !=
+ FJES_DEV_COMMAND_START_DBG_RES_LEN) {
+ result = -ENOMSG;
+ trace_fjes_hw_start_debug_err("Invalid res_buf");
+ } else if (ret == FJES_CMD_STATUS_NORMAL) {
+ switch (res_buf->start_trace.code) {
+ case FJES_CMD_REQ_RES_CODE_NORMAL:
+ result = 0;
+ break;
+ default:
+ result = -EPERM;
+ break;
+ }
+ } else {
+ switch (ret) {
+ case FJES_CMD_STATUS_UNKNOWN:
+ result = -EPERM;
+ break;
+ case FJES_CMD_STATUS_TIMEOUT:
+ trace_fjes_hw_start_debug_err("Busy Timeout");
+ result = -EBUSY;
+ break;
+ case FJES_CMD_STATUS_ERROR_PARAM:
+ case FJES_CMD_STATUS_ERROR_STATUS:
+ default:
+ result = -EPERM;
+ break;
+ }
+ }
+
+ return result;
+}
+
+int fjes_hw_stop_debug(struct fjes_hw *hw)
+{
+ union fjes_device_command_req *req_buf = hw->hw_info.req_buf;
+ union fjes_device_command_res *res_buf = hw->hw_info.res_buf;
+ enum fjes_dev_command_response_e ret;
+ int result = 0;
+
+ if (!hw->hw_info.trace)
+ return -EPERM;
+
+ memset(req_buf, 0, hw->hw_info.req_buf_size);
+ memset(res_buf, 0, hw->hw_info.res_buf_size);
+ req_buf->stop_trace.length = FJES_DEV_COMMAND_STOP_DBG_REQ_LEN;
+
+ res_buf->stop_trace.length = 0;
+ res_buf->stop_trace.code = 0;
+
+ ret = fjes_hw_issue_request_command(hw, FJES_CMD_REQ_STOP_DEBUG);
+ trace_fjes_hw_stop_debug(res_buf);
+
+ if (res_buf->stop_trace.length != FJES_DEV_COMMAND_STOP_DBG_RES_LEN) {
+ trace_fjes_hw_stop_debug_err("Invalid res_buf");
+ result = -ENOMSG;
+ } else if (ret == FJES_CMD_STATUS_NORMAL) {
+ switch (res_buf->stop_trace.code) {
+ case FJES_CMD_REQ_RES_CODE_NORMAL:
+ result = 0;
+ hw->debug_mode = 0;
+ break;
+ default:
+ result = -EPERM;
+ break;
+ }
+ } else {
+ switch (ret) {
+ case FJES_CMD_STATUS_UNKNOWN:
+ result = -EPERM;
+ break;
+ case FJES_CMD_STATUS_TIMEOUT:
+ result = -EBUSY;
+ trace_fjes_hw_stop_debug_err("Busy Timeout");
+ break;
+ case FJES_CMD_STATUS_ERROR_PARAM:
+ case FJES_CMD_STATUS_ERROR_STATUS:
+ default:
+ result = -EPERM;
+ break;
+ }
+ }
+
+ return result;
+}
diff --git a/drivers/net/fjes/fjes_hw.h b/drivers/net/fjes/fjes_hw.h
index 205182e..3a6da09 100644
--- a/drivers/net/fjes/fjes_hw.h
+++ b/drivers/net/fjes/fjes_hw.h
@@ -33,6 +33,9 @@ struct fjes_hw;
#define EP_BUFFER_SUPPORT_VLAN_MAX 4
#define EP_BUFFER_INFO_SIZE 4096
+#define FJES_DEBUG_PAGE_SIZE 4096
+#define FJES_DEBUG_BUFFER_SIZE (16 * FJES_DEBUG_PAGE_SIZE)
+
#define FJES_DEVICE_RESET_TIMEOUT ((17 + 1) * 3 * 8) /* sec */
#define FJES_COMMAND_REQ_TIMEOUT ((5 + 1) * 3 * 8) /* sec */
#define FJES_COMMAND_REQ_BUFF_TIMEOUT (60 * 3) /* sec */
@@ -94,6 +97,12 @@ struct fjes_hw;
#define FJES_DEV_RES_BUF_SIZE(maxep) \
FJES_DEV_COMMAND_INFO_RES_LEN(maxep)
+#define FJES_DEV_COMMAND_START_DBG_REQ_LEN(byte) \
+ (16 + (8 * (byte) / FJES_DEBUG_PAGE_SIZE))
+#define FJES_DEV_COMMAND_START_DBG_RES_LEN (8)
+#define FJES_DEV_COMMAND_STOP_DBG_REQ_LEN (4)
+#define FJES_DEV_COMMAND_STOP_DBG_RES_LEN (8)
+
/* Frame & MTU */
struct esmem_frame {
__le32 frame_size;
@@ -173,6 +182,8 @@ enum fjes_dev_command_request_type {
FJES_CMD_REQ_INFO = 0x0001,
FJES_CMD_REQ_SHARE_BUFFER = 0x0002,
FJES_CMD_REQ_UNSHARE_BUFFER = 0x0004,
+ FJES_CMD_REQ_START_DEBUG = 0x0100,
+ FJES_CMD_REQ_STOP_DEBUG = 0x0200,
};
/* parameter for command control */
@@ -321,6 +332,8 @@ struct fjes_hw {
struct fjes_hw_info hw_info;
spinlock_t rx_status_lock; /* spinlock for rx_status */
+
+ u32 debug_mode;
};
int fjes_hw_init(struct fjes_hw *);
@@ -353,4 +366,6 @@ void *fjes_hw_epbuf_rx_curpkt_get_addr(struct epbuf_handler *, size_t *);
void fjes_hw_epbuf_rx_curpkt_drop(struct epbuf_handler *);
int fjes_hw_epbuf_tx_pkt_send(struct epbuf_handler *, void *, size_t);
+int fjes_hw_start_debug(struct fjes_hw *);
+int fjes_hw_stop_debug(struct fjes_hw *);
#endif /* FJES_HW_H_ */
diff --git a/drivers/net/fjes/fjes_trace.h b/drivers/net/fjes/fjes_trace.h
index 1f0a489..22e8f1a 100644
--- a/drivers/net/fjes/fjes_trace.h
+++ b/drivers/net/fjes/fjes_trace.h
@@ -209,6 +209,75 @@ TRACE_EVENT(fjes_hw_unregister_buff_addr_err,
TP_printk("%s", __get_str(err))
);
+TRACE_EVENT(fjes_hw_start_debug_req,
+ TP_PROTO(union fjes_device_command_req *req_buf),
+ TP_ARGS(req_buf),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, mode)
+ __field(phys_addr_t, buffer)
+ ),
+ TP_fast_assign(
+ __entry->length = req_buf->start_trace.length;
+ __entry->mode = req_buf->start_trace.mode;
+ __entry->buffer = req_buf->start_trace.buffer[0];
+ ),
+ TP_printk("req_buf=[length=%d, mode=%d, buffer=%p]",
+ __entry->length, __entry->mode, (void *)__entry->buffer)
+);
+
+TRACE_EVENT(fjes_hw_start_debug,
+ TP_PROTO(union fjes_device_command_res *res_buf),
+ TP_ARGS(res_buf),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, code)
+ ),
+ TP_fast_assign(
+ __entry->length = res_buf->start_trace.length;
+ __entry->code = res_buf->start_trace.code;
+ ),
+ TP_printk("res_buf=[length=%d, code=%d]", __entry->length, __entry->code)
+);
+
+TRACE_EVENT(fjes_hw_start_debug_err,
+ TP_PROTO(char *err),
+ TP_ARGS(err),
+ TP_STRUCT__entry(
+ __string(err, err)
+ ),
+ TP_fast_assign(
+ __assign_str(err, err)
+ ),
+ TP_printk("%s", __get_str(err))
+);
+
+TRACE_EVENT(fjes_hw_stop_debug,
+ TP_PROTO(union fjes_device_command_res *res_buf),
+ TP_ARGS(res_buf),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, code)
+ ),
+ TP_fast_assign(
+ __entry->length = res_buf->stop_trace.length;
+ __entry->code = res_buf->stop_trace.code;
+ ),
+ TP_printk("res_buf=[length=%d, code=%d]", __entry->length, __entry->code)
+);
+
+TRACE_EVENT(fjes_hw_stop_debug_err,
+ TP_PROTO(char *err),
+ TP_ARGS(err),
+ TP_STRUCT__entry(
+ __string(err, err)
+ ),
+ TP_fast_assign(
+ __assign_str(err, err)
+ ),
+ TP_printk("%s", __get_str(err))
+);
+
/* tracepoints for fjes_main.c */
TRACE_EVENT(fjes_txrx_stop_req_irq_pre,
--
2.6.6
^ permalink raw reply related
* [PATCH net-next v2 3/6] fjes: Add tracepoints in fjes driver
From: Taku Izumi @ 2016-10-14 11:27 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
In-Reply-To: <1476444345-28673-1-git-send-email-izumi.taku@jp.fujitsu.com>
This patch adds tracepoints in fjes driver.
This is useful for debugging purpose.
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
---
drivers/net/fjes/Makefile | 2 +-
drivers/net/fjes/fjes_hw.c | 25 +++-
drivers/net/fjes/fjes_main.c | 5 +
drivers/net/fjes/fjes_trace.c | 30 ++++
drivers/net/fjes/fjes_trace.h | 311 ++++++++++++++++++++++++++++++++++++++++++
5 files changed, 369 insertions(+), 4 deletions(-)
create mode 100644 drivers/net/fjes/fjes_trace.c
create mode 100644 drivers/net/fjes/fjes_trace.h
diff --git a/drivers/net/fjes/Makefile b/drivers/net/fjes/Makefile
index 523e3d7..6705d1b 100644
--- a/drivers/net/fjes/Makefile
+++ b/drivers/net/fjes/Makefile
@@ -27,4 +27,4 @@
obj-$(CONFIG_FUJITSU_ES) += fjes.o
-fjes-objs := fjes_main.o fjes_hw.o fjes_ethtool.o
+fjes-objs := fjes_main.o fjes_hw.o fjes_ethtool.o fjes_trace.o
diff --git a/drivers/net/fjes/fjes_hw.c b/drivers/net/fjes/fjes_hw.c
index 82b56e8..dba59dc 100644
--- a/drivers/net/fjes/fjes_hw.c
+++ b/drivers/net/fjes/fjes_hw.c
@@ -21,6 +21,7 @@
#include "fjes_hw.h"
#include "fjes.h"
+#include "fjes_trace.h"
static void fjes_hw_update_zone_task(struct work_struct *);
static void fjes_hw_epstop_task(struct work_struct *);
@@ -371,7 +372,7 @@ fjes_hw_issue_request_command(struct fjes_hw *hw,
enum fjes_dev_command_response_e ret = FJES_CMD_STATUS_UNKNOWN;
union REG_CR cr;
union REG_CS cs;
- int timeout;
+ int timeout = FJES_COMMAND_REQ_TIMEOUT * 1000;
cr.reg = 0;
cr.bits.req_start = 1;
@@ -408,6 +409,8 @@ fjes_hw_issue_request_command(struct fjes_hw *hw,
}
}
+ trace_fjes_hw_issue_request_command(&cr, &cs, timeout, ret);
+
return ret;
}
@@ -427,11 +430,13 @@ int fjes_hw_request_info(struct fjes_hw *hw)
res_buf->info.code = 0;
ret = fjes_hw_issue_request_command(hw, FJES_CMD_REQ_INFO);
+ trace_fjes_hw_request_info(hw, res_buf);
result = 0;
if (FJES_DEV_COMMAND_INFO_RES_LEN((*hw->hw_info.max_epid)) !=
res_buf->info.length) {
+ trace_fjes_hw_request_info_err("Invalid res_buf");
result = -ENOMSG;
} else if (ret == FJES_CMD_STATUS_NORMAL) {
switch (res_buf->info.code) {
@@ -448,6 +453,7 @@ int fjes_hw_request_info(struct fjes_hw *hw)
result = -EPERM;
break;
case FJES_CMD_STATUS_TIMEOUT:
+ trace_fjes_hw_request_info_err("Timeout");
result = -EBUSY;
break;
case FJES_CMD_STATUS_ERROR_PARAM:
@@ -512,6 +518,8 @@ int fjes_hw_register_buff_addr(struct fjes_hw *hw, int dest_epid,
res_buf->share_buffer.length = 0;
res_buf->share_buffer.code = 0;
+ trace_fjes_hw_register_buff_addr_req(req_buf, buf_pair);
+
ret = fjes_hw_issue_request_command(hw, FJES_CMD_REQ_SHARE_BUFFER);
timeout = FJES_COMMAND_REQ_BUFF_TIMEOUT * 1000;
@@ -532,16 +540,20 @@ int fjes_hw_register_buff_addr(struct fjes_hw *hw, int dest_epid,
result = 0;
+ trace_fjes_hw_register_buff_addr(res_buf, timeout);
+
if (res_buf->share_buffer.length !=
- FJES_DEV_COMMAND_SHARE_BUFFER_RES_LEN)
+ FJES_DEV_COMMAND_SHARE_BUFFER_RES_LEN) {
+ trace_fjes_hw_register_buff_addr_err("Invalid res_buf");
result = -ENOMSG;
- else if (ret == FJES_CMD_STATUS_NORMAL) {
+ } else if (ret == FJES_CMD_STATUS_NORMAL) {
switch (res_buf->share_buffer.code) {
case FJES_CMD_REQ_RES_CODE_NORMAL:
result = 0;
set_bit(dest_epid, &hw->hw_info.buffer_share_bit);
break;
case FJES_CMD_REQ_RES_CODE_BUSY:
+ trace_fjes_hw_register_buff_addr_err("Busy Timeout");
result = -EBUSY;
break;
default:
@@ -554,6 +566,7 @@ int fjes_hw_register_buff_addr(struct fjes_hw *hw, int dest_epid,
result = -EPERM;
break;
case FJES_CMD_STATUS_TIMEOUT:
+ trace_fjes_hw_register_buff_addr_err("Timeout");
result = -EBUSY;
break;
case FJES_CMD_STATUS_ERROR_PARAM:
@@ -595,6 +608,7 @@ int fjes_hw_unregister_buff_addr(struct fjes_hw *hw, int dest_epid)
res_buf->unshare_buffer.length = 0;
res_buf->unshare_buffer.code = 0;
+ trace_fjes_hw_unregister_buff_addr_req(req_buf);
ret = fjes_hw_issue_request_command(hw, FJES_CMD_REQ_UNSHARE_BUFFER);
timeout = FJES_COMMAND_REQ_BUFF_TIMEOUT * 1000;
@@ -616,8 +630,11 @@ int fjes_hw_unregister_buff_addr(struct fjes_hw *hw, int dest_epid)
result = 0;
+ trace_fjes_hw_unregister_buff_addr(res_buf, timeout);
+
if (res_buf->unshare_buffer.length !=
FJES_DEV_COMMAND_UNSHARE_BUFFER_RES_LEN) {
+ trace_fjes_hw_unregister_buff_addr_err("Invalid res_buf");
result = -ENOMSG;
} else if (ret == FJES_CMD_STATUS_NORMAL) {
switch (res_buf->unshare_buffer.code) {
@@ -626,6 +643,7 @@ int fjes_hw_unregister_buff_addr(struct fjes_hw *hw, int dest_epid)
clear_bit(dest_epid, &hw->hw_info.buffer_share_bit);
break;
case FJES_CMD_REQ_RES_CODE_BUSY:
+ trace_fjes_hw_unregister_buff_addr_err("Busy Timeout");
result = -EBUSY;
break;
default:
@@ -638,6 +656,7 @@ int fjes_hw_unregister_buff_addr(struct fjes_hw *hw, int dest_epid)
result = -EPERM;
break;
case FJES_CMD_STATUS_TIMEOUT:
+ trace_fjes_hw_unregister_buff_addr_err("Timeout");
result = -EBUSY;
break;
case FJES_CMD_STATUS_ERROR_PARAM:
diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c
index 03e2383..4c135ae 100644
--- a/drivers/net/fjes/fjes_main.c
+++ b/drivers/net/fjes/fjes_main.c
@@ -27,6 +27,7 @@
#include <linux/interrupt.h>
#include "fjes.h"
+#include "fjes_trace.h"
#define MAJ 1
#define MIN 1
@@ -904,6 +905,7 @@ static void fjes_txrx_stop_req_irq(struct fjes_adapter *adapter,
unsigned long flags;
status = fjes_hw_get_partner_ep_status(hw, src_epid);
+ trace_fjes_txrx_stop_req_irq_pre(hw, src_epid, status);
switch (status) {
case EP_PARTNER_UNSHARE:
case EP_PARTNER_COMPLETE:
@@ -934,6 +936,7 @@ static void fjes_txrx_stop_req_irq(struct fjes_adapter *adapter,
}
break;
}
+ trace_fjes_txrx_stop_req_irq_post(hw, src_epid);
}
static void fjes_stop_req_irq(struct fjes_adapter *adapter, int src_epid)
@@ -945,6 +948,7 @@ static void fjes_stop_req_irq(struct fjes_adapter *adapter, int src_epid)
set_bit(src_epid, &hw->hw_info.buffer_unshare_reserve_bit);
status = fjes_hw_get_partner_ep_status(hw, src_epid);
+ trace_fjes_stop_req_irq_pre(hw, src_epid, status);
switch (status) {
case EP_PARTNER_WAITING:
spin_lock_irqsave(&hw->rx_status_lock, flags);
@@ -968,6 +972,7 @@ static void fjes_stop_req_irq(struct fjes_adapter *adapter, int src_epid)
queue_work(adapter->control_wq, &hw->epstop_task);
break;
}
+ trace_fjes_stop_req_irq_post(hw, src_epid);
}
static void fjes_update_zone_irq(struct fjes_adapter *adapter,
diff --git a/drivers/net/fjes/fjes_trace.c b/drivers/net/fjes/fjes_trace.c
new file mode 100644
index 0000000..066fa76
--- /dev/null
+++ b/drivers/net/fjes/fjes_trace.c
@@ -0,0 +1,30 @@
+/*
+ * FUJITSU Extended Socket Network Device driver
+ * Copyright (c) 2015-2016 FUJITSU LIMITED
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ */
+
+#include <linux/module.h>
+
+#ifndef __CHECKER__
+#include "fjes_hw.h"
+
+#define CREATE_TRACE_POINTS
+#include "fjes_trace.h"
+
+#endif /* __CHECKER__ */
diff --git a/drivers/net/fjes/fjes_trace.h b/drivers/net/fjes/fjes_trace.h
new file mode 100644
index 0000000..1f0a489
--- /dev/null
+++ b/drivers/net/fjes/fjes_trace.h
@@ -0,0 +1,311 @@
+/*
+ * FUJITSU Extended Socket Network Device driver
+ * Copyright (c) 2015-2016 FUJITSU LIMITED
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ */
+
+#if !defined(FJES_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
+#define FJES_TRACE_H_
+
+#include <linux/types.h>
+#include <linux/tracepoint.h>
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM fjes
+
+/* tracepoints for fjes_hw.c */
+
+TRACE_EVENT(fjes_hw_issue_request_command,
+ TP_PROTO(union REG_CR *cr, union REG_CS *cs, int timeout,
+ enum fjes_dev_command_response_e ret),
+ TP_ARGS(cr, cs, timeout, ret),
+ TP_STRUCT__entry(
+ __field(u16, cr_req)
+ __field(u8, cr_error)
+ __field(u16, cr_err_info)
+ __field(u8, cr_req_start)
+ __field(u16, cs_req)
+ __field(u8, cs_busy)
+ __field(u8, cs_complete)
+ __field(int, timeout)
+ __field(int, ret);
+ ),
+ TP_fast_assign(
+ __entry->cr_req = cr->bits.req_code;
+ __entry->cr_error = cr->bits.error;
+ __entry->cr_err_info = cr->bits.err_info;
+ __entry->cr_req_start = cr->bits.req_start;
+ __entry->cs_req = cs->bits.req_code;
+ __entry->cs_busy = cs->bits.busy;
+ __entry->cs_complete = cs->bits.complete;
+ __entry->timeout = timeout;
+ __entry->ret = ret;
+ ),
+ TP_printk("CR=[req=%04x, error=%u, err_info=%04x, req_start=%u], CS=[req=%04x, busy=%u, complete=%u], timeout=%d, ret=%d",
+ __entry->cr_req, __entry->cr_error, __entry->cr_err_info,
+ __entry->cr_req_start, __entry->cs_req, __entry->cs_busy,
+ __entry->cs_complete, __entry->timeout, __entry->ret)
+);
+
+TRACE_EVENT(fjes_hw_request_info,
+ TP_PROTO(struct fjes_hw *hw, union fjes_device_command_res *res_buf),
+ TP_ARGS(hw, res_buf),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, code)
+ __dynamic_array(u8, zone, hw->max_epid)
+ __dynamic_array(u8, status, hw->max_epid)
+ ),
+ TP_fast_assign(
+ int x;
+
+ __entry->length = res_buf->info.length;
+ __entry->code = res_buf->info.code;
+ for (x = 0; x < hw->max_epid; x++) {
+ *((u8 *)__get_dynamic_array(zone) + x) =
+ res_buf->info.info[x].zone;
+ *((u8 *)__get_dynamic_array(status) + x) =
+ res_buf->info.info[x].es_status;
+ }
+ ),
+ TP_printk("res_buf=[length=%d, code=%d, es_zones=%s, es_status=%s]",
+ __entry->length, __entry->code,
+ __print_array(__get_dynamic_array(zone),
+ __get_dynamic_array_len(zone) / sizeof(u8),
+ sizeof(u8)),
+ __print_array(__get_dynamic_array(status),
+ __get_dynamic_array_len(status) / sizeof(u8),
+ sizeof(u8)))
+);
+
+TRACE_EVENT(fjes_hw_request_info_err,
+ TP_PROTO(char *err),
+ TP_ARGS(err),
+ TP_STRUCT__entry(
+ __string(err, err)
+ ),
+ TP_fast_assign(
+ __assign_str(err, err);
+ ),
+ TP_printk("%s", __get_str(err))
+);
+
+TRACE_EVENT(fjes_hw_register_buff_addr_req,
+ TP_PROTO(union fjes_device_command_req *req_buf,
+ struct ep_share_mem_info *buf_pair),
+ TP_ARGS(req_buf, buf_pair),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, epid)
+ __field(u64, tx)
+ __field(size_t, tx_size)
+ __field(u64, rx)
+ __field(size_t, rx_size)
+ ),
+ TP_fast_assign(
+ void *tx, *rx;
+
+ tx = (void *)buf_pair->tx.buffer;
+ rx = (void *)buf_pair->rx.buffer;
+ __entry->length = req_buf->share_buffer.length;
+ __entry->epid = req_buf->share_buffer.epid;
+ __entry->tx_size = buf_pair->tx.size;
+ __entry->rx_size = buf_pair->rx.size;
+ __entry->tx = page_to_phys(vmalloc_to_page(tx)) +
+ offset_in_page(tx);
+ __entry->rx = page_to_phys(vmalloc_to_page(rx)) +
+ offset_in_page(rx);
+ ),
+ TP_printk("req_buf=[length=%d, epid=%d], TX=[phy=0x%016llx, size=%zu], RX=[phy=0x%016llx, size=%zu]",
+ __entry->length, __entry->epid, __entry->tx, __entry->tx_size,
+ __entry->rx, __entry->rx_size)
+);
+
+TRACE_EVENT(fjes_hw_register_buff_addr,
+ TP_PROTO(union fjes_device_command_res *res_buf, int timeout),
+ TP_ARGS(res_buf, timeout),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, code)
+ __field(int, timeout)
+ ),
+ TP_fast_assign(
+ __entry->length = res_buf->share_buffer.length;
+ __entry->code = res_buf->share_buffer.code;
+ __entry->timeout = timeout;
+ ),
+ TP_printk("res_buf=[length=%d, code=%d], timeout=%d",
+ __entry->length, __entry->code, __entry->timeout)
+);
+
+TRACE_EVENT(fjes_hw_register_buff_addr_err,
+ TP_PROTO(char *err),
+ TP_ARGS(err),
+ TP_STRUCT__entry(
+ __string(err, err)
+ ),
+ TP_fast_assign(
+ __assign_str(err, err);
+ ),
+ TP_printk("%s", __get_str(err))
+);
+
+TRACE_EVENT(fjes_hw_unregister_buff_addr_req,
+ TP_PROTO(union fjes_device_command_req *req_buf),
+ TP_ARGS(req_buf),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, epid)
+ ),
+ TP_fast_assign(
+ __entry->length = req_buf->unshare_buffer.length;
+ __entry->epid = req_buf->unshare_buffer.epid;
+ ),
+ TP_printk("req_buf=[length=%d, epid=%d]",
+ __entry->length, __entry->epid)
+);
+
+TRACE_EVENT(fjes_hw_unregister_buff_addr,
+ TP_PROTO(union fjes_device_command_res *res_buf, int timeout),
+ TP_ARGS(res_buf, timeout),
+ TP_STRUCT__entry(
+ __field(int, length)
+ __field(int, code)
+ __field(int, timeout)
+ ),
+ TP_fast_assign(
+ __entry->length = res_buf->unshare_buffer.length;
+ __entry->code = res_buf->unshare_buffer.code;
+ __entry->timeout = timeout;
+ ),
+ TP_printk("res_buf=[length=%d, code=%d], timeout=%d",
+ __entry->length, __entry->code, __entry->timeout)
+);
+
+TRACE_EVENT(fjes_hw_unregister_buff_addr_err,
+ TP_PROTO(char *err),
+ TP_ARGS(err),
+ TP_STRUCT__entry(
+ __string(err, err)
+ ),
+ TP_fast_assign(
+ __assign_str(err, err);
+ ),
+ TP_printk("%s", __get_str(err))
+);
+
+/* tracepoints for fjes_main.c */
+
+TRACE_EVENT(fjes_txrx_stop_req_irq_pre,
+ TP_PROTO(struct fjes_hw *hw, int src_epid,
+ enum ep_partner_status status),
+ TP_ARGS(hw, src_epid, status),
+ TP_STRUCT__entry(
+ __field(int, src_epid)
+ __field(enum ep_partner_status, status)
+ __field(u8, ep_status)
+ __field(unsigned long, txrx_stop_req_bit)
+ __field(u16, rx_status)
+ ),
+ TP_fast_assign(
+ __entry->src_epid = src_epid;
+ __entry->status = status;
+ __entry->ep_status = hw->hw_info.share->ep_status[src_epid];
+ __entry->txrx_stop_req_bit = hw->txrx_stop_req_bit;
+ __entry->rx_status =
+ hw->ep_shm_info[src_epid].tx.info->v1i.rx_status;
+ ),
+ TP_printk("epid=%d, partner_status=%d, ep_status=%x, txrx_stop_req_bit=%016lx, tx.rx_status=%08x",
+ __entry->src_epid, __entry->status, __entry->ep_status,
+ __entry->txrx_stop_req_bit, __entry->rx_status)
+);
+
+TRACE_EVENT(fjes_txrx_stop_req_irq_post,
+ TP_PROTO(struct fjes_hw *hw, int src_epid),
+ TP_ARGS(hw, src_epid),
+ TP_STRUCT__entry(
+ __field(int, src_epid)
+ __field(u8, ep_status)
+ __field(unsigned long, txrx_stop_req_bit)
+ __field(u16, rx_status)
+ ),
+ TP_fast_assign(
+ __entry->src_epid = src_epid;
+ __entry->ep_status = hw->hw_info.share->ep_status[src_epid];
+ __entry->txrx_stop_req_bit = hw->txrx_stop_req_bit;
+ __entry->rx_status = hw->ep_shm_info[src_epid].tx.info->v1i.rx_status;
+ ),
+ TP_printk("epid=%d, ep_status=%x, txrx_stop_req_bit=%016lx, tx.rx_status=%08x",
+ __entry->src_epid, __entry->ep_status,
+ __entry->txrx_stop_req_bit, __entry->rx_status)
+);
+
+TRACE_EVENT(fjes_stop_req_irq_pre,
+ TP_PROTO(struct fjes_hw *hw, int src_epid,
+ enum ep_partner_status status),
+ TP_ARGS(hw, src_epid, status),
+ TP_STRUCT__entry(
+ __field(int, src_epid)
+ __field(enum ep_partner_status, status)
+ __field(u8, ep_status)
+ __field(unsigned long, txrx_stop_req_bit)
+ __field(u16, rx_status)
+ ),
+ TP_fast_assign(
+ __entry->src_epid = src_epid;
+ __entry->status = status;
+ __entry->ep_status = hw->hw_info.share->ep_status[src_epid];
+ __entry->txrx_stop_req_bit = hw->txrx_stop_req_bit;
+ __entry->rx_status =
+ hw->ep_shm_info[src_epid].tx.info->v1i.rx_status;
+ ),
+ TP_printk("epid=%d, partner_status=%d, ep_status=%x, txrx_stop_req_bit=%016lx, tx.rx_status=%08x",
+ __entry->src_epid, __entry->status, __entry->ep_status,
+ __entry->txrx_stop_req_bit, __entry->rx_status)
+);
+
+TRACE_EVENT(fjes_stop_req_irq_post,
+ TP_PROTO(struct fjes_hw *hw, int src_epid),
+ TP_ARGS(hw, src_epid),
+ TP_STRUCT__entry(
+ __field(int, src_epid)
+ __field(u8, ep_status)
+ __field(unsigned long, txrx_stop_req_bit)
+ __field(u16, rx_status)
+ ),
+ TP_fast_assign(
+ __entry->src_epid = src_epid;
+ __entry->ep_status = hw->hw_info.share->ep_status[src_epid];
+ __entry->txrx_stop_req_bit = hw->txrx_stop_req_bit;
+ __entry->rx_status =
+ hw->ep_shm_info[src_epid].tx.info->v1i.rx_status;
+ ),
+ TP_printk("epid=%d, ep_status=%x, txrx_stop_req_bit=%016lx, tx.rx_status=%08x",
+ __entry->src_epid, __entry->ep_status,
+ __entry->txrx_stop_req_bit, __entry->rx_status)
+);
+
+#endif /* FJES_TRACE_H_ */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH ../../../drivers/net/fjes
+#define TRACE_INCLUDE_FILE fjes_trace
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
--
2.6.6
^ permalink raw reply related
* [PATCH net-next v2 2/6] fjes: Enhance ethtool -S for fjes driver
From: Taku Izumi @ 2016-10-14 11:27 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
In-Reply-To: <1476444345-28673-1-git-send-email-izumi.taku@jp.fujitsu.com>
This patch enhances ethtool -S for fjes driver so that
EP related statistics can be retrieved.
The following statistics can be displayed via ethtool -S:
ep%d_com_regist_buf_exec
ep%d_com_unregist_buf_exec
ep%d_send_intr_rx
ep%d_send_intr_unshare
ep%d_send_intr_zoneupdate
ep%d_recv_intr_rx
ep%d_recv_intr_unshare
ep%d_recv_intr_stop
ep%d_recv_intr_zoneupdate
ep%d_tx_buffer_full
ep%d_tx_dropped_not_shared
ep%d_tx_dropped_ver_mismatch
ep%d_tx_dropped_buf_size_mismatch
ep%d_tx_dropped_vlanid_mismatch
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
---
drivers/net/fjes/fjes_ethtool.c | 70 ++++++++++++++++++++++++++++++++++++++++-
drivers/net/fjes/fjes_hw.c | 9 ++++++
drivers/net/fjes/fjes_hw.h | 19 +++++++++++
drivers/net/fjes/fjes_main.c | 44 +++++++++++++++++++++++---
4 files changed, 137 insertions(+), 5 deletions(-)
diff --git a/drivers/net/fjes/fjes_ethtool.c b/drivers/net/fjes/fjes_ethtool.c
index 8397634..68ef287 100644
--- a/drivers/net/fjes/fjes_ethtool.c
+++ b/drivers/net/fjes/fjes_ethtool.c
@@ -49,10 +49,18 @@ static const struct fjes_stats fjes_gstrings_stats[] = {
FJES_STAT("tx_dropped", stats64.tx_dropped),
};
+#define FJES_EP_STATS_LEN 14
+#define FJES_STATS_LEN \
+ (ARRAY_SIZE(fjes_gstrings_stats) + \
+ ((&((struct fjes_adapter *)netdev_priv(netdev))->hw)->max_epid - 1) * \
+ FJES_EP_STATS_LEN)
+
static void fjes_get_ethtool_stats(struct net_device *netdev,
struct ethtool_stats *stats, u64 *data)
{
struct fjes_adapter *adapter = netdev_priv(netdev);
+ struct fjes_hw *hw = &adapter->hw;
+ int epidx;
char *p;
int i;
@@ -61,11 +69,39 @@ static void fjes_get_ethtool_stats(struct net_device *netdev,
data[i] = (fjes_gstrings_stats[i].sizeof_stat == sizeof(u64))
? *(u64 *)p : *(u32 *)p;
}
+ for (epidx = 0; epidx < hw->max_epid; epidx++) {
+ if (epidx == hw->my_epid)
+ continue;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .com_regist_buf_exec;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .com_unregist_buf_exec;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats.send_intr_rx;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats.send_intr_unshare;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .send_intr_zoneupdate;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats.recv_intr_rx;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats.recv_intr_unshare;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats.recv_intr_stop;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .recv_intr_zoneupdate;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats.tx_buffer_full;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .tx_dropped_not_shared;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .tx_dropped_ver_mismatch;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .tx_dropped_buf_size_mismatch;
+ data[i++] = hw->ep_shm_info[epidx].ep_stats
+ .tx_dropped_vlanid_mismatch;
+ }
}
static void fjes_get_strings(struct net_device *netdev,
u32 stringset, u8 *data)
{
+ struct fjes_adapter *adapter = netdev_priv(netdev);
+ struct fjes_hw *hw = &adapter->hw;
u8 *p = data;
int i;
@@ -76,6 +112,38 @@ static void fjes_get_strings(struct net_device *netdev,
ETH_GSTRING_LEN);
p += ETH_GSTRING_LEN;
}
+ for (i = 0; i < hw->max_epid; i++) {
+ if (i == hw->my_epid)
+ continue;
+ sprintf(p, "ep%u_com_regist_buf_exec", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_com_unregist_buf_exec", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_send_intr_rx", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_send_intr_unshare", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_send_intr_zoneupdate", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_recv_intr_rx", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_recv_intr_unshare", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_recv_intr_stop", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_recv_intr_zoneupdate", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_tx_buffer_full", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_tx_dropped_not_shared", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_tx_dropped_ver_mismatch", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_tx_dropped_buf_size_mismatch", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "ep%u_tx_dropped_vlanid_mismatch", i);
+ p += ETH_GSTRING_LEN;
+ }
break;
}
}
@@ -84,7 +152,7 @@ static int fjes_get_sset_count(struct net_device *netdev, int sset)
{
switch (sset) {
case ETH_SS_STATS:
- return ARRAY_SIZE(fjes_gstrings_stats);
+ return FJES_STATS_LEN;
default:
return -EOPNOTSUPP;
}
diff --git a/drivers/net/fjes/fjes_hw.c b/drivers/net/fjes/fjes_hw.c
index 0dbafed..82b56e8 100644
--- a/drivers/net/fjes/fjes_hw.c
+++ b/drivers/net/fjes/fjes_hw.c
@@ -752,6 +752,7 @@ void fjes_hw_raise_epstop(struct fjes_hw *hw)
case EP_PARTNER_SHARED:
fjes_hw_raise_interrupt(hw, epidx,
REG_ICTL_MASK_TXRX_STOP_REQ);
+ hw->ep_shm_info[epidx].ep_stats.send_intr_unshare += 1;
break;
default:
break;
@@ -1062,6 +1063,9 @@ static void fjes_hw_update_zone_task(struct work_struct *work)
break;
}
mutex_unlock(&hw->hw_info.lock);
+
+ hw->ep_shm_info[epidx].ep_stats
+ .com_regist_buf_exec += 1;
}
if (test_bit(epidx, &unshare_bit)) {
@@ -1085,6 +1089,9 @@ static void fjes_hw_update_zone_task(struct work_struct *work)
mutex_unlock(&hw->hw_info.lock);
+ hw->ep_shm_info[epidx].ep_stats
+ .com_unregist_buf_exec += 1;
+
if (ret == 0) {
spin_lock_irqsave(&hw->rx_status_lock, flags);
fjes_hw_setup_epbuf(
@@ -1099,6 +1106,8 @@ static void fjes_hw_update_zone_task(struct work_struct *work)
fjes_hw_raise_interrupt(hw, epidx,
REG_ICTL_MASK_TXRX_STOP_REQ);
+ hw->ep_shm_info[epidx].ep_stats.send_intr_unshare += 1;
+
set_bit(epidx, &hw->txrx_stop_req_bit);
spin_lock_irqsave(&hw->rx_status_lock, flags);
hw->ep_shm_info[epidx].tx.
diff --git a/drivers/net/fjes/fjes_hw.h b/drivers/net/fjes/fjes_hw.h
index 1445ac9..205182e 100644
--- a/drivers/net/fjes/fjes_hw.h
+++ b/drivers/net/fjes/fjes_hw.h
@@ -228,6 +228,24 @@ union ep_buffer_info {
};
+/* statistics of EP */
+struct fjes_drv_ep_stats {
+ u64 com_regist_buf_exec;
+ u64 com_unregist_buf_exec;
+ u64 send_intr_rx;
+ u64 send_intr_unshare;
+ u64 send_intr_zoneupdate;
+ u64 recv_intr_rx;
+ u64 recv_intr_unshare;
+ u64 recv_intr_stop;
+ u64 recv_intr_zoneupdate;
+ u64 tx_buffer_full;
+ u64 tx_dropped_not_shared;
+ u64 tx_dropped_ver_mismatch;
+ u64 tx_dropped_buf_size_mismatch;
+ u64 tx_dropped_vlanid_mismatch;
+};
+
/* buffer pair for Extended Partition */
struct ep_share_mem_info {
struct epbuf_handler {
@@ -238,6 +256,7 @@ struct ep_share_mem_info {
} tx, rx;
struct rtnl_link_stats64 net_stats;
+ struct fjes_drv_ep_stats ep_stats;
u16 tx_status_work;
diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c
index e46b1eb..03e2383 100644
--- a/drivers/net/fjes/fjes_main.c
+++ b/drivers/net/fjes/fjes_main.c
@@ -366,6 +366,8 @@ static int fjes_setup_resources(struct fjes_adapter *adapter)
FJES_ZONING_STATUS_ENABLE)) {
fjes_hw_raise_interrupt(hw, epidx,
REG_ICTL_MASK_INFO_UPDATE);
+ hw->ep_shm_info[epidx].ep_stats
+ .send_intr_zoneupdate += 1;
}
}
@@ -397,6 +399,9 @@ static int fjes_setup_resources(struct fjes_adapter *adapter)
adapter->force_reset = true;
return result;
}
+
+ hw->ep_shm_info[epidx].ep_stats
+ .com_regist_buf_exec += 1;
}
}
@@ -422,6 +427,8 @@ static void fjes_free_resources(struct fjes_adapter *adapter)
result = fjes_hw_unregister_buff_addr(hw, epidx);
mutex_unlock(&hw->hw_info.lock);
+ hw->ep_shm_info[epidx].ep_stats.com_unregist_buf_exec += 1;
+
if (result)
reset_flag = true;
@@ -567,6 +574,7 @@ static void fjes_raise_intr_rxdata_task(struct work_struct *work)
FJES_RX_POLL_WORK)) {
fjes_hw_raise_interrupt(hw, epid,
REG_ICTL_MASK_RX_DATA);
+ hw->ep_shm_info[epid].ep_stats.send_intr_rx += 1;
}
}
@@ -663,6 +671,9 @@ fjes_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
pstatus = fjes_hw_get_partner_ep_status(hw, dest_epid);
if (pstatus != EP_PARTNER_SHARED) {
+ if (!is_multi)
+ hw->ep_shm_info[dest_epid].ep_stats
+ .tx_dropped_not_shared += 1;
ret = NETDEV_TX_OK;
} else if (!fjes_hw_check_epbuf_version(
&adapter->hw.ep_shm_info[dest_epid].rx, 0)) {
@@ -670,6 +681,8 @@ fjes_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
adapter->stats64.tx_carrier_errors += 1;
hw->ep_shm_info[dest_epid].net_stats
.tx_carrier_errors += 1;
+ hw->ep_shm_info[dest_epid].ep_stats
+ .tx_dropped_ver_mismatch += 1;
ret = NETDEV_TX_OK;
} else if (!fjes_hw_check_mtu(
@@ -679,12 +692,16 @@ fjes_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
hw->ep_shm_info[dest_epid].net_stats.tx_dropped += 1;
adapter->stats64.tx_errors += 1;
hw->ep_shm_info[dest_epid].net_stats.tx_errors += 1;
+ hw->ep_shm_info[dest_epid].ep_stats
+ .tx_dropped_buf_size_mismatch += 1;
ret = NETDEV_TX_OK;
} else if (vlan &&
!fjes_hw_check_vlan_id(
&adapter->hw.ep_shm_info[dest_epid].rx,
vlan_id)) {
+ hw->ep_shm_info[dest_epid].ep_stats
+ .tx_dropped_vlanid_mismatch += 1;
ret = NETDEV_TX_OK;
} else {
if (len < VLAN_ETH_HLEN) {
@@ -718,6 +735,8 @@ fjes_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
ret = NETDEV_TX_OK;
} else {
netif_trans_update(netdev);
+ hw->ep_shm_info[dest_epid].ep_stats
+ .tx_buffer_full += 1;
netif_tx_stop_queue(cur_queue);
if (!work_pending(&adapter->tx_stall_task))
@@ -970,21 +989,33 @@ static irqreturn_t fjes_intr(int irq, void *data)
icr = fjes_hw_capture_interrupt_status(hw);
if (icr & REG_IS_MASK_IS_ASSERT) {
- if (icr & REG_ICTL_MASK_RX_DATA)
+ if (icr & REG_ICTL_MASK_RX_DATA) {
fjes_rx_irq(adapter, icr & REG_IS_MASK_EPID);
+ hw->ep_shm_info[icr & REG_IS_MASK_EPID].ep_stats
+ .recv_intr_rx += 1;
+ }
- if (icr & REG_ICTL_MASK_DEV_STOP_REQ)
+ if (icr & REG_ICTL_MASK_DEV_STOP_REQ) {
fjes_stop_req_irq(adapter, icr & REG_IS_MASK_EPID);
+ hw->ep_shm_info[icr & REG_IS_MASK_EPID].ep_stats
+ .recv_intr_stop += 1;
+ }
- if (icr & REG_ICTL_MASK_TXRX_STOP_REQ)
+ if (icr & REG_ICTL_MASK_TXRX_STOP_REQ) {
fjes_txrx_stop_req_irq(adapter, icr & REG_IS_MASK_EPID);
+ hw->ep_shm_info[icr & REG_IS_MASK_EPID].ep_stats
+ .recv_intr_unshare += 1;
+ }
if (icr & REG_ICTL_MASK_TXRX_STOP_DONE)
fjes_hw_set_irqmask(hw,
REG_ICTL_MASK_TXRX_STOP_DONE, true);
- if (icr & REG_ICTL_MASK_INFO_UPDATE)
+ if (icr & REG_ICTL_MASK_INFO_UPDATE) {
fjes_update_zone_irq(adapter, icr & REG_IS_MASK_EPID);
+ hw->ep_shm_info[icr & REG_IS_MASK_EPID].ep_stats
+ .recv_intr_zoneupdate += 1;
+ }
ret = IRQ_HANDLED;
} else {
@@ -1364,6 +1395,8 @@ static void fjes_watch_unshare_task(struct work_struct *work)
break;
}
mutex_unlock(&hw->hw_info.lock);
+ hw->ep_shm_info[epidx].ep_stats
+ .com_unregist_buf_exec += 1;
spin_lock_irqsave(&hw->rx_status_lock, flags);
fjes_hw_setup_epbuf(&hw->ep_shm_info[epidx].tx,
@@ -1406,6 +1439,9 @@ static void fjes_watch_unshare_task(struct work_struct *work)
}
mutex_unlock(&hw->hw_info.lock);
+ hw->ep_shm_info[epidx].ep_stats
+ .com_unregist_buf_exec += 1;
+
spin_lock_irqsave(&hw->rx_status_lock, flags);
fjes_hw_setup_epbuf(
&hw->ep_shm_info[epidx].tx,
--
2.6.6
^ permalink raw reply related
* [PATCH net-next v2 1/6] fjes: ethtool -d support for fjes driver
From: Taku Izumi @ 2016-10-14 11:27 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
In-Reply-To: <1476444345-28673-1-git-send-email-izumi.taku@jp.fujitsu.com>
This patch adds implementation of supporting
ethtool -d for fjes driver. By using ethtool -d,
you can get registers dump of Exetnded socket device.
# ethtool -d es0
Offset Values
------ ------
0x0000: 01 00 00 00 08 00 00 00 00 00 00 00 00 00 00 00
0x0010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0020: 02 00 00 80 02 00 00 80 64 a6 58 08 07 00 00 00
0x0030: 00 00 00 00 28 80 00 00 00 00 f9 e3 06 00 00 00
0x0040: 00 00 00 00 18 00 00 00 80 a4 58 08 07 00 00 00
0x0050: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0060: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0080: 00 00 00 00 00 00 e0 7f 00 00 01 00 00 00 01 00
0x0090: 00 00 00 00
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
---
drivers/net/fjes/fjes_ethtool.c | 48 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 48 insertions(+)
diff --git a/drivers/net/fjes/fjes_ethtool.c b/drivers/net/fjes/fjes_ethtool.c
index 9c218e1..8397634 100644
--- a/drivers/net/fjes/fjes_ethtool.c
+++ b/drivers/net/fjes/fjes_ethtool.c
@@ -121,12 +121,60 @@ static int fjes_get_settings(struct net_device *netdev,
return 0;
}
+static int fjes_get_regs_len(struct net_device *netdev)
+{
+#define FJES_REGS_LEN 37
+ return FJES_REGS_LEN * sizeof(u32);
+}
+
+static void fjes_get_regs(struct net_device *netdev,
+ struct ethtool_regs *regs, void *p)
+{
+ struct fjes_adapter *adapter = netdev_priv(netdev);
+ struct fjes_hw *hw = &adapter->hw;
+ u32 *regs_buff = p;
+
+ memset(p, 0, FJES_REGS_LEN * sizeof(u32));
+
+ regs->version = 1;
+
+ /* Information registers */
+ regs_buff[0] = rd32(XSCT_OWNER_EPID);
+ regs_buff[1] = rd32(XSCT_MAX_EP);
+
+ /* Device Control registers */
+ regs_buff[4] = rd32(XSCT_DCTL);
+
+ /* Command Control registers */
+ regs_buff[8] = rd32(XSCT_CR);
+ regs_buff[9] = rd32(XSCT_CS);
+ regs_buff[10] = rd32(XSCT_SHSTSAL);
+ regs_buff[11] = rd32(XSCT_SHSTSAH);
+
+ regs_buff[13] = rd32(XSCT_REQBL);
+ regs_buff[14] = rd32(XSCT_REQBAL);
+ regs_buff[15] = rd32(XSCT_REQBAH);
+
+ regs_buff[17] = rd32(XSCT_RESPBL);
+ regs_buff[18] = rd32(XSCT_RESPBAL);
+ regs_buff[19] = rd32(XSCT_RESPBAH);
+
+ /* Interrupt Control registers */
+ regs_buff[32] = rd32(XSCT_IS);
+ regs_buff[33] = rd32(XSCT_IMS);
+ regs_buff[34] = rd32(XSCT_IMC);
+ regs_buff[35] = rd32(XSCT_IG);
+ regs_buff[36] = rd32(XSCT_ICTL);
+}
+
static const struct ethtool_ops fjes_ethtool_ops = {
.get_settings = fjes_get_settings,
.get_drvinfo = fjes_get_drvinfo,
.get_ethtool_stats = fjes_get_ethtool_stats,
.get_strings = fjes_get_strings,
.get_sset_count = fjes_get_sset_count,
+ .get_regs = fjes_get_regs,
+ .get_regs_len = fjes_get_regs_len,
};
void fjes_set_ethtool_ops(struct net_device *netdev)
--
2.6.6
^ permalink raw reply related
* [PATCH net-next v2 0/6] FUJITSU Extended Socket driver version 1.2
From: Taku Izumi @ 2016-10-14 11:25 UTC (permalink / raw)
To: netdev, davem; +Cc: Taku Izumi
This patchset updates FUJITSU Extended Socket network driver into version 1.2.
This includes the following enhancements:
- ethtool -d support
- ethtool -S enhancement
- ethtool -w/-W support
- Add some debugging feature (tracepoints etc)
v1 -> v2:
- Use u64 instead of phys_addr_t as TP_STRUCT__entry
- Use ethtool facility to achieve debug mode instead of using debugfs
Taku Izumi (6):
fjes: ethtool -d support for fjes driver
fjes: Enhance ethtool -S for fjes driver
fjes: Add tracepoints in fjes driver
fjes: ethtool -w and -W support for fjes driver
fjes: Add debugfs entry for EP status information in fjes driver
fjes: Update fjes driver version : 1.2
drivers/net/fjes/Makefile | 2 +-
drivers/net/fjes/fjes.h | 16 ++
drivers/net/fjes/fjes_debugfs.c | 117 +++++++++++++
drivers/net/fjes/fjes_ethtool.c | 181 ++++++++++++++++++-
drivers/net/fjes/fjes_hw.c | 171 +++++++++++++++++-
drivers/net/fjes/fjes_hw.h | 34 ++++
drivers/net/fjes/fjes_main.c | 63 ++++++-
drivers/net/fjes/fjes_trace.c | 30 ++++
drivers/net/fjes/fjes_trace.h | 380 ++++++++++++++++++++++++++++++++++++++++
9 files changed, 983 insertions(+), 11 deletions(-)
create mode 100644 drivers/net/fjes/fjes_debugfs.c
create mode 100644 drivers/net/fjes/fjes_trace.c
create mode 100644 drivers/net/fjes/fjes_trace.h
--
2.6.6
^ permalink raw reply
* 731540 netdev
From: ppsalome.khum @ 2016-10-14 11:35 UTC (permalink / raw)
To: netdev
[-- Attachment #1: EMAIL_94157035_netdev.zip --]
[-- Type: application/zip, Size: 858 bytes --]
^ permalink raw reply
* RE: [PATCH v2 net-next 0/7] qed*: driver updates
From: Chopra, Manish @ 2016-10-14 9:54 UTC (permalink / raw)
To: Manish Chopra, davem@davemloft.net
Cc: netdev@vger.kernel.org, Yuval.Mintz@qlogic.com
In-Reply-To: <1476357771-1578-1-git-send-email-manish.chopra@qlogic.com>
> -----Original Message-----
> From: Manish Chopra [mailto:manish.chopra@qlogic.com]
> Sent: Thursday, October 13, 2016 4:53 PM
> To: davem@davemloft.net
> Cc: netdev@vger.kernel.org; Yuval.Mintz@qlogic.com; Chopra, Manish
> <Manish.Chopra@cavium.com>
> Subject: [PATCH v2 net-next 0/7] qed*: driver updates
>
> From: Manish Chopra <manish.chopra@caviumnetworks.com>
>
> Hi David,
>
> There are several new additions in this series;
> Most are connected to either Tx offloading or Rx classifications
> [either fastpath changes or supporting configuration].
>
> In addition, there's a single IOV enhancement.
>
> Please consider applying this series to `net-next'.
>
> V2:
> Added a fix for the race in ramrod handling
> pointed by Eric Dumazet [patch 7].
>
Hi David, The last patch in this series has caused below kbuild failure.
drivers/net/ethernet/qlogic/qed/qed_spq.c: In function 'qed_spq_blocking_cb':
drivers/net/ethernet/qlogic/qed/qed_spq.c:62:85: error: call to '__compiletime_assert_60' declared with attribute error: Need native word sized stores/loads for atomicity.
I will re-spin the series with kbuild fix and will send out soon.
Thanks,
Manish
^ permalink raw reply
* Re: [PATCH net 2/2] conntrack: enable to tune gc parameters
From: Florian Westphal @ 2016-10-14 11:16 UTC (permalink / raw)
To: Pablo Neira Ayuso
Cc: Florian Westphal, Nicolas Dichtel, davem, netdev, netfilter-devel
In-Reply-To: <20161014105327.GA8655@salvia>
Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> I would prefer not to expose sysctl knobs, if we don't really know
> what good default values are good, then we cannot expect our users to
> know this for us.
>
> I would go tune this in a way that this resembles to the previous
> behaviour.
I do not see how this is possible without reverting to old per-conntrack
timer scheme.
With per-ct timer userspace gets notified the moment the timer
fires, without it notification comes 'when kernel detects the timeout'
which in worst case, as Nicholas describes, is when gc worker comes
along.
You can run the gc worker every jiffie of course, but thats just
wasting cpu cycles (and you still get a small delay).
I don't see a way to do run-time tuning except faster restarts when
old entries start accumulating. This is what the code tries to do,
perhaps you have a better idea for the 'next gc run' computation.
^ permalink raw reply
* Re: net/sctp: BUG: KASAN: stack-out-of-bounds in memcmp
From: Xin Long @ 2016-10-14 11:13 UTC (permalink / raw)
To: Baozeng Ding
Cc: Vladislav Yasevich, Neil Horman, David Miller, linux-sctp,
network dev
In-Reply-To: <6843fbba-a11c-8bc4-495a-294dc7fdcc35@gmail.com>
On Sat, Aug 20, 2016 at 3:51 PM, Baozeng Ding <sploving1@gmail.com> wrote:
> Hello all,
> The following program triggers stack-out-of-bounds in memcmp. The kernel version is 4.8.0-rc1+ (on Aug 13 commit 118253a593bd1c57de2d1193df1ccffe1abe745b). Thanks.
...
>
> #define _GNU_SOURCE
> #include <unistd.h>
> #include <stdint.h>
> #include <sys/socket.h>
> #include <sys/mman.h>
> #include <linux/in.h>
> #include <fcntl.h>
> #include <string.h>
> #include <stdio.h>
>
> int main()
> {
> int fd;
> mmap((void *)0x20000000ul, 0xff2000ul, 0x3ul, 0x32ul, -1, 0x0ul);
> fd = socket(AF_INET6, SOCK_STREAM, IPPROTO_SCTP);
> memcpy((void*)0x20f82f80, "\x0a\x00\xab\x12\x72\xd4\x19\x9a\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x85\xda\x00\xa0\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00", 128);
> bind(fd, (struct sockaddr*)0x20f82f80ul, 0x80ul);
> *(uint64_t*)0x202e1fc8 = (uint64_t)0x20f77f80;
> *(uint32_t*)0x202e1fd0 = (uint32_t)0x80;
> *(uint64_t*)0x202e1fd8 = (uint64_t)0x20f7dfe0;
> *(uint64_t*)0x202e1fe0 = (uint64_t)0x2;
> *(uint64_t*)0x202e1fe8 = (uint64_t)0x20f77000;
> *(uint64_t*)0x202e1ff0 = (uint64_t)0x3;
> *(uint32_t*)0x202e1ff8 = (uint32_t)0x80;
> memcpy((void*)0x20f77f80, "\x0a\x00\xab\x12\xb0\xb3\x20\x7b\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\xc2\xc2\x0b\xb2\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00", 128);
> *(uint64_t*)0x20f7dfe0 = (uint64_t)0x20f77fc5;
> *(uint64_t*)0x20f7dfe8 = (uint64_t)0x3b;
> *(uint64_t*)0x20f7dff0 = (uint64_t)0x20f77fac;
> *(uint64_t*)0x20f7dff8 = (uint64_t)0x54;
> memcpy((void*)0x20f77fc5, "\xa5\x7d\xf3\xc4\xfe\xd3\xfd\x44\x63\x00\x8c\x1e\x4c\x2e\x8d\x8d\x9a\x9c\x9c\x9d\x5b\x7c\xe1\x06\xf7\x15\x16\xed\x68\xd1\xfc\xf4\xa4\x3a\xe4\x69\x51\x16\x74\xf4\x1a\xcf\x0e\x99\xc3\xa3\x87\xe7\x81\x6c\x10\x78\x75\x17\x69\x9d\x11\x0c\xc7", 59);
> memcpy((void*)0x20f77fac, "\x86\x08\x89\x3c\xf3\x58\xea\xe7\x64\x6a\xfb\xb5\xe8\xdd\x5f\x69\xa5\xd4\xdc\xd9\xe7\x71\x95\x07\x78\x7b\x21\xda\x43\x9c\x62\x4d\xca\x64\xb5\x6e\x96\x55\xe9\x58\x76\x66\x1d\xb9\x7b\xe6\x20\xc1\xa9\xed\x70\xc1\x2b\x7c\x86\x8c\xba\x28\xb3\x2c\xb9\x64\xb7\x84\x65\x0d\x7f\xa6\x98\x6f\x49\xcb\x35\xad\x5a\xdf\x13\x75\x99\x57\x7e\xbb\x38\x89", 84);
> *(uint64_t*)0x20f77000 = (uint64_t)0x15;
> *(uint32_t*)0x20f77008 = (uint32_t)0x1;
> *(uint32_t*)0x20f7700c = (uint32_t)0xfffffffffffffffe;
> *(uint8_t*)0x20f77010 = (uint8_t)0xbb;
> *(uint8_t*)0x20f77011 = (uint8_t)0x2;
> *(uint8_t*)0x20f77012 = (uint8_t)0x5;
> *(uint8_t*)0x20f77013 = (uint8_t)0x2;
> *(uint8_t*)0x20f77014 = (uint8_t)0x80000000;
> *(uint64_t*)0x20f77015 = (uint64_t)0x10;
> *(uint32_t*)0x20f7701d = (uint32_t)0xffff;
> *(uint32_t*)0x20f77021 = (uint32_t)0x1;
> *(uint64_t*)0x20f77025 = (uint64_t)0x13;
> *(uint32_t*)0x20f7702d = (uint32_t)0x6;
> *(uint32_t*)0x20f77031 = (uint32_t)0xfffffffffffffe00;
> *(uint8_t*)0x20f77035 = (uint8_t)0x80000000;
> *(uint8_t*)0x20f77036 = (uint8_t)0xfffffffffffffff8;
> sendmmsg(fd, (struct mmsghdr *)0x202e1fc8ul, 0x1ul, 0x1ul);
> return 0;
> }
>
Hi, Baozeng, I couldn't reproduce this issue with this script,
even in 118253a593bd1c57de2d1193df1ccffe1abe745b
do I need to do some extra config for this ?
^ permalink raw reply
* Re: [mac80211] BUG_ON with current -git (4.8.0-11417-g24532f7)
From: Ard Biesheuvel @ 2016-10-14 11:11 UTC (permalink / raw)
To: Johannes Berg
Cc: Andy Lutomirski, Stephen Rothwell, linux-next@vger.kernel.org,
Sergey Senozhatsky, Network Development, Sergey Senozhatsky,
Herbert Xu, David S. Miller, Linux Wireless List,
linux-kernel@vger.kernel.org, Jouni Malinen
In-Reply-To: <1476439212.31114.37.camel@sipsolutions.net>
On 14 October 2016 at 11:00, Johannes Berg <johannes@sipsolutions.net> wrote:
>
>> So why is the performance hit acceptable for ESP but not for WPA? We
>> could easily implement the same thing, i.e.,
>> kmalloc(GFP_ATOMIC)/kfree the aead_req struct rather than allocate it
>> on the stack
>
> Yeah, maybe we should. It's likely a much bigger allocation, but I
> don't actually know if that affects speed.
>
> In most cases where you want high performance we never hit this anyway
> since we'll have hardware crypto. I know for our (Intel's) devices we
> normally never hit these code paths.
>
> But on the other hand, you also did your changes for a reason, and the
> only reason I can see of that is performance. So you'd be the one with
> most "skin in the game", I guess?
>
Well, what sucks here is that the accelerated driver I implemented for
arm64 does not actually need this, as long as we take aad[] off the
stack. And note that the API was changed since my patch, to add aad[]
to the scatterlist: prior to this change, it used
aead_request_set_assoc() to set the associated data separately.
^ permalink raw reply
* Re: [PATCH net 2/2] conntrack: enable to tune gc parameters
From: Pablo Neira Ayuso @ 2016-10-14 10:53 UTC (permalink / raw)
To: Florian Westphal; +Cc: Nicolas Dichtel, davem, netdev, netfilter-devel
In-Reply-To: <20161014103726.GA10404@breakpoint.cc>
On Fri, Oct 14, 2016 at 12:37:26PM +0200, Florian Westphal wrote:
> Nicolas Dichtel <nicolas.dichtel@6wind.com> wrote:
> > Le 13/10/2016 à 22:43, Florian Westphal a écrit :
[...]
> > > (Or cause too many useless scans)
> > >
> > > Another idea worth trying might be to get rid of the max cap and
> > > instead break early in case too many jiffies expired.
> > >
> > > I don't want to add sysctl knobs for this unless absolutely needed; its already
> > > possible to 'force' eviction cycle by running 'conntrack -L'.
> > >
> > Sure, but this is not a "real" solution, just a workaround.
> > We need to find a way to deliver conntrack deletion events in a reasonable
> > delay, whatever the traffic on the machine is.
>
> Agree, but that depends on what 'reasonable' means and what kind of
> uneeded cpu churn we're willing to add.
>
> We can add a sysctl for this but we should use a low default to not do
> too much unneeded work.
>
> So what about your original patch, but only add
>
> nf_conntrack_gc_interval
>
> (and also add instant-resched in case entire budget was consumed)?
I would prefer not to expose sysctl knobs, if we don't really know
what good default values are good, then we cannot expect our users to
know this for us.
I would go tune this in a way that this resembles to the previous
behaviour.
^ permalink raw reply
* Re: [PATCH net 4/5] net/ncsi: Choose hot channel as active one if necessary
From: Gavin Shan @ 2016-10-14 10:39 UTC (permalink / raw)
To: Joel Stanley; +Cc: Gavin Shan, netdev, davem
In-Reply-To: <CACPK8XfPxORiE91uyNVivC9pw2dP+gnfzB5WUeGiGBKc-KTtEg@mail.gmail.com>
On Fri, Oct 14, 2016 at 04:32:36PM +1030, Joel Stanley wrote:
>On Fri, Oct 14, 2016 at 1:23 PM, Gavin Shan <gwshan@linux.vnet.ibm.com> wrote:
>> The issue was found on BCM5718 which has two NCSI channels in one
>> package: C0 and C1. C0 is in link-up state while C1 is in link-down
>> state. C0 is chosen as active channel until unplugging and plugging
>> C0's cable: On unplugging C0's cable, LSC (Link State Change) AEN
>> packet received on C0 to report link-down event. After that, C1 is
>> chosen as active channel. LSC AEN for link-up event is lost on C0
>> when plugging C0's cable back. We lose the network even C0 is usable.
>
>Why do we lose the LCS AEN packet?
>
>Is this a bug in the BCM5718? If so, we shouldn't put it in the common
>ncsi code without adding a quirk for that hardware.
>
It's not a BCM5718 bug. LSC AEN is only received on the active channel.
After the failover (C0 -> C1), C0 becomes inactive and LSC AEN packet
won't be received on it as expected.
Thanks,
Gavin
>>
>> This resolves the issue by recording the (hot) channel that was ever
>> chosen as active one. The hot channel is chosen to be active one
>> if none of available channels in link-up state. With this, C0 is still
>> the active one after unplugging C0's cable. LSC AEN packet received
>> on C0 when plugging its cable back.
>>
>> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
>> ---
>> net/ncsi/internal.h | 1 +
>> net/ncsi/ncsi-manage.c | 22 +++++++++++++++++++---
>> 2 files changed, 20 insertions(+), 3 deletions(-)
>>
>> diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h
>> index eac4858..1308a56 100644
>> --- a/net/ncsi/internal.h
>> +++ b/net/ncsi/internal.h
>> @@ -265,6 +265,7 @@ struct ncsi_dev_priv {
>> #endif
>> unsigned int package_num; /* Number of packages */
>> struct list_head packages; /* List of packages */
>> + struct ncsi_channel *hot_channel; /* Channel was ever active */
>> struct ncsi_request requests[256]; /* Request table */
>> unsigned int request_id; /* Last used request ID */
>> #define NCSI_REQ_START_IDX 1
>> diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c
>> index e959979..cccedcf 100644
>> --- a/net/ncsi/ncsi-manage.c
>> +++ b/net/ncsi/ncsi-manage.c
>> @@ -625,6 +625,7 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp)
>> struct net_device *dev = nd->dev;
>> struct ncsi_package *np = ndp->active_package;
>> struct ncsi_channel *nc = ndp->active_channel;
>> + struct ncsi_channel *hot_nc = NULL;
>> struct ncsi_cmd_arg nca;
>> unsigned char index;
>> unsigned long flags;
>> @@ -730,12 +731,20 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp)
>> break;
>> case ncsi_dev_state_config_done:
>> spin_lock_irqsave(&nc->lock, flags);
>> - if (nc->modes[NCSI_MODE_LINK].data[2] & 0x1)
>> + if (nc->modes[NCSI_MODE_LINK].data[2] & 0x1) {
>> + hot_nc = nc;
>> nc->state = NCSI_CHANNEL_ACTIVE;
>> - else
>> + } else {
>> + hot_nc = NULL;
>> nc->state = NCSI_CHANNEL_INACTIVE;
>> + }
>> spin_unlock_irqrestore(&nc->lock, flags);
>>
>> + /* Update the hot channel */
>> + spin_lock_irqsave(&ndp->lock, flags);
>> + ndp->hot_channel = hot_nc;
>> + spin_unlock_irqrestore(&ndp->lock, flags);
>> +
>> ncsi_start_channel_monitor(nc);
>> ncsi_process_next_channel(ndp);
>> break;
>> @@ -753,10 +762,14 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp)
>> static int ncsi_choose_active_channel(struct ncsi_dev_priv *ndp)
>> {
>> struct ncsi_package *np;
>> - struct ncsi_channel *nc, *found;
>> + struct ncsi_channel *nc, *found, *hot_nc;
>> struct ncsi_channel_mode *ncm;
>> unsigned long flags;
>>
>> + spin_lock_irqsave(&ndp->lock, flags);
>> + hot_nc = ndp->hot_channel;
>> + spin_unlock_irqrestore(&ndp->lock, flags);
>> +
>> /* The search is done once an inactive channel with up
>> * link is found.
>> */
>> @@ -774,6 +787,9 @@ static int ncsi_choose_active_channel(struct ncsi_dev_priv *ndp)
>> if (!found)
>> found = nc;
>>
>> + if (nc == hot_nc)
>> + found = nc;
>> +
>> ncm = &nc->modes[NCSI_MODE_LINK];
>> if (ncm->data[2] & 0x1) {
>> spin_unlock_irqrestore(&nc->lock, flags);
>> --
>> 2.1.0
>>
>
^ permalink raw reply
* Re: [PATCH net 2/2] conntrack: enable to tune gc parameters
From: Florian Westphal @ 2016-10-14 10:37 UTC (permalink / raw)
To: Nicolas Dichtel; +Cc: Florian Westphal, davem, pablo, netdev, netfilter-devel
In-Reply-To: <f1df137d-7278-f6fb-3237-1bd20afadb48@6wind.com>
Nicolas Dichtel <nicolas.dichtel@6wind.com> wrote:
> Le 13/10/2016 à 22:43, Florian Westphal a écrit :
> > Nicolas Dichtel <nicolas.dichtel@6wind.com> wrote:
> >> Le 10/10/2016 à 16:04, Florian Westphal a écrit :
> >>> Nicolas Dichtel <nicolas.dichtel@6wind.com> wrote:
> >>>> After commit b87a2f9199ea ("netfilter: conntrack: add gc worker to remove
> >>>> timed-out entries"), netlink conntrack deletion events may be sent with a
> >>>> huge delay. It could be interesting to let the user tweak gc parameters
> >>>> depending on its use case.
> >>>
> >>> Hmm, care to elaborate?
> >>>
> >>> I am not against doing this but I'd like to hear/read your use case.
> >>>
> >>> The expectation is that in almot all cases eviction will happen from
> >>> packet path. The gc worker is jusdt there for case where a busy system
> >>> goes idle.
> >> It was precisely that case. After a period of activity, the event is sent a long
> >> time after the timeout. If the router does not manage a lot of flows, why not
> >> trying to parse more entries instead of the default 1/64 of the table?
> >> In fact, I don't understand why using GC_MAX_BUCKETS_DIV instead of using always
> >> GC_MAX_BUCKETS whatever the size of the table is.
> >
> > I wanted to make sure that we have a known upper bound on the number of
> > buckets we process so that we do not block other pending kworker items
> > for too long.
> I don't understand. GC_MAX_BUCKETS is the upper bound and I agree that it is
> needed. But why GC_MAX_BUCKETS_DIV (ie 1/64)?
> In other words, why this line:
> goal = min(nf_conntrack_htable_size / GC_MAX_BUCKETS_DIV, GC_MAX_BUCKETS);
> instead of:
> goal = GC_MAX_BUCKETS;
Sure, we can do that. But why is a fixed size better than a fraction?
E.g. with 8k buckets and simple goal = GC_MAX_BUCKETS we scan entire
table on every run, currently we only scan 128.
I wanted to keep too many destroy notifications from firing at once
but maybe i was too paranoid...
> > (Or cause too many useless scans)
> >
> > Another idea worth trying might be to get rid of the max cap and
> > instead break early in case too many jiffies expired.
> >
> > I don't want to add sysctl knobs for this unless absolutely needed; its already
> > possible to 'force' eviction cycle by running 'conntrack -L'.
> >
> Sure, but this is not a "real" solution, just a workaround.
> We need to find a way to deliver conntrack deletion events in a reasonable
> delay, whatever the traffic on the machine is.
Agree, but that depends on what 'reasonable' means and what kind of
uneeded cpu churn we're willing to add.
We can add a sysctl for this but we should use a low default to not do
too much unneeded work.
So what about your original patch, but only add
nf_conntrack_gc_interval
(and also add instant-resched in case entire budget was consumed)?
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox